/[pcre]/code/trunk/pcre_internal.h
ViewVC logotype

Diff of /code/trunk/pcre_internal.h

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 268 by ph10, Thu Nov 15 10:28:09 2007 UTC revision 1425 by ph10, Tue Dec 31 17:44:40 2013 UTC
# Line 7  Line 7 
7  and semantics are as close as possible to those of the Perl 5 language.  and semantics are as close as possible to those of the Perl 5 language.
8    
9                         Written by Philip Hazel                         Written by Philip Hazel
10             Copyright (c) 1997-2007 University of Cambridge             Copyright (c) 1997-2013 University of Cambridge
11    
12  -----------------------------------------------------------------------------  -----------------------------------------------------------------------------
13  Redistribution and use in source and binary forms, with or without  Redistribution and use in source and binary forms, with or without
# Line 40  POSSIBILITY OF SUCH DAMAGE. Line 40  POSSIBILITY OF SUCH DAMAGE.
40    
41  /* This header contains definitions that are shared between the different  /* This header contains definitions that are shared between the different
42  modules, but which are not relevant to the exported API. This includes some  modules, but which are not relevant to the exported API. This includes some
43  functions whose names all begin with "_pcre_". */  functions whose names all begin with "_pcre_", "_pcre16_" or "_pcre32_"
44    depending on the PRIV macro. */
45    
46  #ifndef PCRE_INTERNAL_H  #ifndef PCRE_INTERNAL_H
47  #define PCRE_INTERNAL_H  #define PCRE_INTERNAL_H
48    
49  /* Define DEBUG to get debugging output on stdout. */  /* Define PCRE_DEBUG to get debugging output on stdout. */
50    
51  #if 0  #if 0
52  #define DEBUG  #define PCRE_DEBUG
53    #endif
54    
55    /* PCRE is compiled as an 8 bit library if it is not requested otherwise. */
56    
57    #if !defined COMPILE_PCRE16 && !defined COMPILE_PCRE32
58    #define COMPILE_PCRE8
59    #endif
60    
61    /* If SUPPORT_UCP is defined, SUPPORT_UTF must also be defined. The
62    "configure" script ensures this, but not everybody uses "configure". */
63    
64    #if defined SUPPORT_UCP && !(defined SUPPORT_UTF)
65    #define SUPPORT_UTF 1
66    #endif
67    
68    /* We define SUPPORT_UTF if SUPPORT_UTF8 is enabled for compatibility
69    reasons with existing code. */
70    
71    #if defined SUPPORT_UTF8 && !(defined SUPPORT_UTF)
72    #define SUPPORT_UTF 1
73    #endif
74    
75    /* Fixme: SUPPORT_UTF8 should be eventually disappear from the code.
76    Until then we define it if SUPPORT_UTF is defined. */
77    
78    #if defined SUPPORT_UTF && !(defined SUPPORT_UTF8)
79    #define SUPPORT_UTF8 1
80    #endif
81    
82    /* We do not support both EBCDIC and UTF-8/16/32 at the same time. The "configure"
83    script prevents both being selected, but not everybody uses "configure". */
84    
85    #if defined EBCDIC && defined SUPPORT_UTF
86    #error The use of both EBCDIC and SUPPORT_UTF is not supported.
87  #endif  #endif
88    
89  /* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef  /* Use a macro for debugging printing, 'cause that eliminates the use of #ifdef
# Line 60  It turns out that the Mac Debugging.h he Line 95  It turns out that the Mac Debugging.h he
95  be absolutely sure we get our version. */  be absolutely sure we get our version. */
96    
97  #undef DPRINTF  #undef DPRINTF
98  #ifdef DEBUG  #ifdef PCRE_DEBUG
99  #define DPRINTF(p) printf p  #define DPRINTF(p) printf p
100  #else  #else
101  #define DPRINTF(p) /* Nothing */  #define DPRINTF(p) /* Nothing */
# Line 72  setjmp and stdarg are used is when NO_RE Line 107  setjmp and stdarg are used is when NO_RE
107    
108  #include <ctype.h>  #include <ctype.h>
109  #include <limits.h>  #include <limits.h>
 #include <setjmp.h>  
 #include <stdarg.h>  
110  #include <stddef.h>  #include <stddef.h>
111  #include <stdio.h>  #include <stdio.h>
112  #include <stdlib.h>  #include <stdlib.h>
113  #include <string.h>  #include <string.h>
114    
115    /* Valgrind (memcheck) support */
116    
117    #ifdef SUPPORT_VALGRIND
118    #include <valgrind/memcheck.h>
119    #endif
120    
121  /* When compiling a DLL for Windows, the exported symbols have to be declared  /* When compiling a DLL for Windows, the exported symbols have to be declared
122  using some MS magic. I found some useful information on this web page:  using some MS magic. I found some useful information on this web page:
123  http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the  http://msdn2.microsoft.com/en-us/library/y4h7bcy6(VS.80).aspx. According to the
# Line 132  PCRE_EXP_DATA_DEFN only if they are not Line 171  PCRE_EXP_DATA_DEFN only if they are not
171  #  endif  #  endif
172  #endif  #endif
173    
174  /* We need to have types that specify unsigned 16-bit and 32-bit integers. We  /* When compiling with the MSVC compiler, it is sometimes necessary to include
175    a "calling convention" before exported function names. (This is secondhand
176    information; I know nothing about MSVC myself). For example, something like
177    
178      void __cdecl function(....)
179    
180    might be needed. In order so make this easy, all the exported functions have
181    PCRE_CALL_CONVENTION just before their names. It is rarely needed; if not
182    set, we ensure here that it has no effect. */
183    
184    #ifndef PCRE_CALL_CONVENTION
185    #define PCRE_CALL_CONVENTION
186    #endif
187    
188    /* We need to have types that specify unsigned 8, 16 and 32-bit integers. We
189  cannot determine these outside the compilation (e.g. by running a program as  cannot determine these outside the compilation (e.g. by running a program as
190  part of "configure") because PCRE is often cross-compiled for use on other  part of "configure") because PCRE is often cross-compiled for use on other
191  systems. Instead we make use of the maximum sizes that are available at  systems. Instead we make use of the maximum sizes that are available at
192  preprocessor time in standard C environments. */  preprocessor time in standard C environments. */
193    
194    typedef unsigned char pcre_uint8;
195    
196  #if USHRT_MAX == 65535  #if USHRT_MAX == 65535
197    typedef unsigned short pcre_uint16;  typedef unsigned short pcre_uint16;
198    typedef short pcre_int16;
199    #define PCRE_UINT16_MAX USHRT_MAX
200    #define PCRE_INT16_MAX SHRT_MAX
201  #elif UINT_MAX == 65535  #elif UINT_MAX == 65535
202    typedef unsigned int pcre_uint16;  typedef unsigned int pcre_uint16;
203    typedef int pcre_int16;
204    #define PCRE_UINT16_MAX UINT_MAX
205    #define PCRE_INT16_MAX INT_MAX
206  #else  #else
207    #error Cannot determine a type for 16-bit unsigned integers  #error Cannot determine a type for 16-bit integers
208  #endif  #endif
209    
210  #if UINT_MAX == 4294967295  #if UINT_MAX == 4294967295U
211    typedef unsigned int pcre_uint32;  typedef unsigned int pcre_uint32;
212  #elif ULONG_MAX == 4294967295  typedef int pcre_int32;
213    typedef unsigned long int pcre_uint32;  #define PCRE_UINT32_MAX UINT_MAX
214    #define PCRE_INT32_MAX INT_MAX
215    #elif ULONG_MAX == 4294967295UL
216    typedef unsigned long int pcre_uint32;
217    typedef long int pcre_int32;
218    #define PCRE_UINT32_MAX ULONG_MAX
219    #define PCRE_INT32_MAX LONG_MAX
220  #else  #else
221    #error Cannot determine a type for 32-bit unsigned integers  #error Cannot determine a type for 32-bit integers
222    #endif
223    
224    /* When checking for integer overflow in pcre_compile(), we need to handle
225    large integers. If a 64-bit integer type is available, we can use that.
226    Otherwise we have to cast to double, which of course requires floating point
227    arithmetic. Handle this by defining a macro for the appropriate type. If
228    stdint.h is available, include it; it may define INT64_MAX. Systems that do not
229    have stdint.h (e.g. Solaris) may have inttypes.h. The macro int64_t may be set
230    by "configure". */
231    
232    #if defined HAVE_STDINT_H
233    #include <stdint.h>
234    #elif defined HAVE_INTTYPES_H
235    #include <inttypes.h>
236    #endif
237    
238    #if defined INT64_MAX || defined int64_t
239    #define INT64_OR_DOUBLE int64_t
240    #else
241    #define INT64_OR_DOUBLE double
242  #endif  #endif
243    
244  /* All character handling must be done as unsigned characters. Otherwise there  /* All character handling must be done as unsigned characters. Otherwise there
245  are problems with top-bit-set characters and functions such as isspace().  are problems with top-bit-set characters and functions such as isspace().
246  However, we leave the interface to the outside world as char *, because that  However, we leave the interface to the outside world as char * or short *,
247  should make things easier for callers. We define a short type for unsigned char  because that should make things easier for callers. This character type is
248  to save lots of typing. I tried "uchar", but it causes problems on Digital  called pcre_uchar.
249  Unix, where it is defined in sys/types, so use "uschar" instead. */  
250    The IN_UCHARS macro multiply its argument with the byte size of the current
251    pcre_uchar type. Useful for memcpy and such operations, whose require the
252    byte size of their input/output buffers.
253    
254    The MAX_255 macro checks whether its pcre_uchar input is less than 256.
255    
256    The TABLE_GET macro is designed for accessing elements of tables whose contain
257    exactly 256 items. When the character is able to contain more than 256
258    items, some check is needed before accessing these tables.
259    */
260    
261  typedef unsigned char uschar;  #if defined COMPILE_PCRE8
262    
263    typedef unsigned char pcre_uchar;
264    #define IN_UCHARS(x) (x)
265    #define MAX_255(c) 1
266    #define TABLE_GET(c, table, default) ((table)[c])
267    
268    #elif defined COMPILE_PCRE16
269    
270    #if USHRT_MAX != 65535
271    /* This is a warning message. Change PCRE_UCHAR16 to a 16 bit data type in
272    pcre.h(.in) and disable (comment out) this message. */
273    #error Warning: PCRE_UCHAR16 is not a 16 bit data type.
274    #endif
275    
276    typedef pcre_uint16 pcre_uchar;
277    #define UCHAR_SHIFT (1)
278    #define IN_UCHARS(x) ((x) << UCHAR_SHIFT)
279    #define MAX_255(c) ((c) <= 255u)
280    #define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
281    
282    #elif defined COMPILE_PCRE32
283    
284    typedef pcre_uint32 pcre_uchar;
285    #define UCHAR_SHIFT (2)
286    #define IN_UCHARS(x) ((x) << UCHAR_SHIFT)
287    #define MAX_255(c) ((c) <= 255u)
288    #define TABLE_GET(c, table, default) (MAX_255(c)? ((table)[c]):(default))
289    
290    #else
291    #error Unsupported compiling mode
292    #endif /* COMPILE_PCRE[8|16|32] */
293    
294  /* This is an unsigned int value that no character can ever have. UTF-8  /* This is an unsigned int value that no character can ever have. UTF-8
295  characters only go up to 0x7fffffff (though Unicode doesn't go beyond  characters only go up to 0x7fffffff (though Unicode doesn't go beyond
# Line 184  start/end of string field names are. */ Line 312  start/end of string field names are. */
312  #define IS_NEWLINE(p) \  #define IS_NEWLINE(p) \
313    ((NLBLOCK->nltype != NLTYPE_FIXED)? \    ((NLBLOCK->nltype != NLTYPE_FIXED)? \
314      ((p) < NLBLOCK->PSEND && \      ((p) < NLBLOCK->PSEND && \
315       _pcre_is_newline((p), NLBLOCK->nltype, NLBLOCK->PSEND, &(NLBLOCK->nllen),\       PRIV(is_newline)((p), NLBLOCK->nltype, NLBLOCK->PSEND, \
316         utf8)) \         &(NLBLOCK->nllen), utf)) \
317      : \      : \
318      ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \      ((p) <= NLBLOCK->PSEND - NLBLOCK->nllen && \
319       (p)[0] == NLBLOCK->nl[0] && \       *p == NLBLOCK->nl[0] && (NLBLOCK->nllen == 1 || p[1] == NLBLOCK->nl[1]) \
      (NLBLOCK->nllen == 1 || (p)[1] == NLBLOCK->nl[1]) \  
320      ) \      ) \
321    )    )
322    
# Line 198  start/end of string field names are. */ Line 325  start/end of string field names are. */
325  #define WAS_NEWLINE(p) \  #define WAS_NEWLINE(p) \
326    ((NLBLOCK->nltype != NLTYPE_FIXED)? \    ((NLBLOCK->nltype != NLTYPE_FIXED)? \
327      ((p) > NLBLOCK->PSSTART && \      ((p) > NLBLOCK->PSSTART && \
328       _pcre_was_newline((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \       PRIV(was_newline)((p), NLBLOCK->nltype, NLBLOCK->PSSTART, \
329         &(NLBLOCK->nllen), utf8)) \         &(NLBLOCK->nllen), utf)) \
330      : \      : \
331      ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \      ((p) >= NLBLOCK->PSSTART + NLBLOCK->nllen && \
332       (p)[-NLBLOCK->nllen] == NLBLOCK->nl[0] && \       *(p - NLBLOCK->nllen) == NLBLOCK->nl[0] &&              \
333       (NLBLOCK->nllen == 1 || (p)[-NLBLOCK->nllen+1] == NLBLOCK->nl[1]) \       (NLBLOCK->nllen == 1 || *(p - NLBLOCK->nllen + 1) == NLBLOCK->nl[1]) \
334      ) \      ) \
335    )    )
336    
# Line 217  used for the external interface and appe Line 344  used for the external interface and appe
344  must begin with PCRE_. */  must begin with PCRE_. */
345    
346  #ifdef CUSTOM_SUBJECT_PTR  #ifdef CUSTOM_SUBJECT_PTR
347  #define PCRE_SPTR CUSTOM_SUBJECT_PTR  #define PCRE_PUCHAR CUSTOM_SUBJECT_PTR
 #define USPTR CUSTOM_SUBJECT_PTR  
348  #else  #else
349  #define PCRE_SPTR const char *  #define PCRE_PUCHAR const pcre_uchar *
 #define USPTR const unsigned char *  
350  #endif  #endif
351    
   
   
352  /* Include the public PCRE header and the definitions of UCP character property  /* Include the public PCRE header and the definitions of UCP character property
353  values. */  values. */
354    
355  #include "pcre.h"  #include "pcre.h"
356  #include "ucp.h"  #include "ucp.h"
357    
358    #ifdef COMPILE_PCRE32
359    /* Assert that the public PCRE_UCHAR32 is a 32-bit type */
360    typedef int __assert_pcre_uchar32_size[sizeof(PCRE_UCHAR32) == 4 ? 1 : -1];
361    #endif
362    
363  /* When compiling for use with the Virtual Pascal compiler, these functions  /* When compiling for use with the Virtual Pascal compiler, these functions
364  need to have their names changed. PCRE must be compiled with the -DVPCOMPAT  need to have their names changed. PCRE must be compiled with the -DVPCOMPAT
365  option on the command line. */  option on the command line. */
# Line 293  The macros are controlled by the value o Line 421  The macros are controlled by the value o
421  the config.h file, but can be overridden by using -D on the command line. This  the config.h file, but can be overridden by using -D on the command line. This
422  is automated on Unix systems via the "configure" command. */  is automated on Unix systems via the "configure" command. */
423    
424    #if defined COMPILE_PCRE8
425    
426  #if LINK_SIZE == 2  #if LINK_SIZE == 2
427    
428  #define PUT(a,n,d)   \  #define PUT(a,n,d)   \
# Line 329  is automated on Unix systems via the "co Line 459  is automated on Unix systems via the "co
459  #define GET(a,n) \  #define GET(a,n) \
460    (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])    (((a)[n] << 24) | ((a)[(n)+1] << 16) | ((a)[(n)+2] << 8) | (a)[(n)+3])
461    
462  #define MAX_PATTERN_SIZE (1 << 30)   /* Keep it positive */  /* Keep it positive */
463    #define MAX_PATTERN_SIZE (1 << 30)
464    
465    #else
466    #error LINK_SIZE must be either 2, 3, or 4
467    #endif
468    
469    #elif defined COMPILE_PCRE16
470    
471    #if LINK_SIZE == 2
472    
473    /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
474    #undef LINK_SIZE
475    #define LINK_SIZE 1
476    
477    #define PUT(a,n,d)   \
478      (a[n] = (d))
479    
480    #define GET(a,n) \
481      (a[n])
482    
483    #define MAX_PATTERN_SIZE (1 << 16)
484    
485    #elif LINK_SIZE == 3 || LINK_SIZE == 4
486    
487    /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
488    #undef LINK_SIZE
489    #define LINK_SIZE 2
490    
491    #define PUT(a,n,d)   \
492      (a[n] = (d) >> 16), \
493      (a[(n)+1] = (d) & 65535)
494    
495    #define GET(a,n) \
496      (((a)[n] << 16) | (a)[(n)+1])
497    
498    /* Keep it positive */
499    #define MAX_PATTERN_SIZE (1 << 30)
500    
501  #else  #else
502  #error LINK_SIZE must be either 2, 3, or 4  #error LINK_SIZE must be either 2, 3, or 4
503  #endif  #endif
504    
505    #elif defined COMPILE_PCRE32
506    
507    /* Only supported LINK_SIZE is 4 */
508    /* Redefine LINK_SIZE as a multiple of sizeof(pcre_uchar) */
509    #undef LINK_SIZE
510    #define LINK_SIZE 1
511    
512    #define PUT(a,n,d)   \
513      (a[n] = (d))
514    
515    #define GET(a,n) \
516      (a[n])
517    
518    /* Keep it positive */
519    #define MAX_PATTERN_SIZE (1 << 30)
520    
521    #else
522    #error Unsupported compiling mode
523    #endif /* COMPILE_PCRE[8|16|32] */
524    
525  /* Convenience macro defined in terms of the others */  /* Convenience macro defined in terms of the others */
526    
# Line 346  is automated on Unix systems via the "co Line 531  is automated on Unix systems via the "co
531  offsets changes. There are used for repeat counts and for other things such as  offsets changes. There are used for repeat counts and for other things such as
532  capturing parenthesis numbers in back references. */  capturing parenthesis numbers in back references. */
533    
534    #if defined COMPILE_PCRE8
535    
536    #define IMM2_SIZE 2
537    
538  #define PUT2(a,n,d)   \  #define PUT2(a,n,d)   \
539    a[n] = (d) >> 8; \    a[n] = (d) >> 8; \
540    a[(n)+1] = (d) & 255    a[(n)+1] = (d) & 255
541    
542    /* For reasons that I do not understand, the expression in this GET2 macro is
543    treated by gcc as a signed expression, even when a is declared as unsigned. It
544    seems that any kind of arithmetic results in a signed value. */
545    
546  #define GET2(a,n) \  #define GET2(a,n) \
547    (((a)[n] << 8) | (a)[(n)+1])    (unsigned int)(((a)[n] << 8) | (a)[(n)+1])
548    
549  #define PUT2INC(a,n,d)  PUT2(a,n,d), a += 2  #elif defined COMPILE_PCRE16
550    
551    #define IMM2_SIZE 1
552    
553  /* When UTF-8 encoding is being used, a character is no longer just a single  #define PUT2(a,n,d)   \
554  byte. The macros for character handling generate simple sequences when used in     a[n] = d
555  byte-mode, and more complicated ones for UTF-8 characters. BACKCHAR should  
556  never be called in byte mode. To make sure it can never even appear when UTF-8  #define GET2(a,n) \
557  support is omitted, we don't even define it. */     a[n]
558    
559    #elif defined COMPILE_PCRE32
560    
561    #define IMM2_SIZE 1
562    
563    #define PUT2(a,n,d)   \
564       a[n] = d
565    
566    #define GET2(a,n) \
567       a[n]
568    
569    #else
570    #error Unsupported compiling mode
571    #endif /* COMPILE_PCRE[8|16|32] */
572    
573    #define PUT2INC(a,n,d)  PUT2(a,n,d), a += IMM2_SIZE
574    
575  #ifndef SUPPORT_UTF8  /* The maximum length of a MARK name is currently one data unit; it may be
576    changed in future to be a fixed number of bytes or to depend on LINK_SIZE. */
577    
578    #if defined COMPILE_PCRE16 || defined COMPILE_PCRE32
579    #define MAX_MARK ((1u << 16) - 1)
580    #else
581    #define MAX_MARK ((1u << 8) - 1)
582    #endif
583    
584    /* When UTF encoding is being used, a character is no longer just a single
585    byte. The macros for character handling generate simple sequences when used in
586    character-mode, and more complicated ones for UTF characters. GETCHARLENTEST
587    and other macros are not used when UTF is not supported, so they are not
588    defined. To make sure they can never even appear when UTF support is omitted,
589    we don't even define them. */
590    
591    #ifndef SUPPORT_UTF
592    
593    /* #define MAX_VALUE_FOR_SINGLE_CHAR */
594    /* #define HAS_EXTRALEN(c) */
595    /* #define GET_EXTRALEN(c) */
596    /* #define NOT_FIRSTCHAR(c) */
597  #define GETCHAR(c, eptr) c = *eptr;  #define GETCHAR(c, eptr) c = *eptr;
598  #define GETCHARTEST(c, eptr) c = *eptr;  #define GETCHARTEST(c, eptr) c = *eptr;
599  #define GETCHARINC(c, eptr) c = *eptr++;  #define GETCHARINC(c, eptr) c = *eptr++;
600  #define GETCHARINCTEST(c, eptr) c = *eptr++;  #define GETCHARINCTEST(c, eptr) c = *eptr++;
601  #define GETCHARLEN(c, eptr, len) c = *eptr;  #define GETCHARLEN(c, eptr, len) c = *eptr;
602    /* #define GETCHARLENTEST(c, eptr, len) */
603  /* #define BACKCHAR(eptr) */  /* #define BACKCHAR(eptr) */
604    /* #define FORWARDCHAR(eptr) */
605    /* #define ACROSSCHAR(condition, eptr, action) */
606    
607  #else   /* SUPPORT_UTF8 */  #else   /* SUPPORT_UTF */
608    
609  /* Get the next UTF-8 character, not advancing the pointer. This is called when  /* Tests whether the code point needs extra characters to decode. */
 we know we are in UTF-8 mode. */  
610    
611  #define GETCHAR(c, eptr) \  #define HASUTF8EXTRALEN(c) ((c) >= 0xc0)
612    c = *eptr; \  
613    if (c >= 0xc0) \  /* Base macro to pick up the remaining bytes of a UTF-8 character, not
614    advancing the pointer. */
615    
616    #define GETUTF8(c, eptr) \
617      { \      { \
618      int gcii; \      if ((c & 0x20) == 0) \
619      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \        c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
620      int gcss = 6*gcaa; \      else if ((c & 0x10) == 0) \
621      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \        c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
622      for (gcii = 1; gcii <= gcaa; gcii++) \      else if ((c & 0x08) == 0) \
623          c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
624          ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
625        else if ((c & 0x04) == 0) \
626          c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
627              ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
628              (eptr[4] & 0x3f); \
629        else \
630          c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
631              ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
632              ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
633        }
634    
635    /* Base macro to pick up the remaining bytes of a UTF-8 character, advancing
636    the pointer. */
637    
638    #define GETUTF8INC(c, eptr) \
639        { \
640        if ((c & 0x20) == 0) \
641          c = ((c & 0x1f) << 6) | (*eptr++ & 0x3f); \
642        else if ((c & 0x10) == 0) \
643          { \
644          c = ((c & 0x0f) << 12) | ((*eptr & 0x3f) << 6) | (eptr[1] & 0x3f); \
645          eptr += 2; \
646          } \
647        else if ((c & 0x08) == 0) \
648          { \
649          c = ((c & 0x07) << 18) | ((*eptr & 0x3f) << 12) | \
650              ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
651          eptr += 3; \
652          } \
653        else if ((c & 0x04) == 0) \
654          { \
655          c = ((c & 0x03) << 24) | ((*eptr & 0x3f) << 18) | \
656              ((eptr[1] & 0x3f) << 12) | ((eptr[2] & 0x3f) << 6) | \
657              (eptr[3] & 0x3f); \
658          eptr += 4; \
659          } \
660        else \
661        { \        { \
662        gcss -= 6; \        c = ((c & 0x01) << 30) | ((*eptr & 0x3f) << 24) | \
663        c |= (eptr[gcii] & 0x3f) << gcss; \            ((eptr[1] & 0x3f) << 18) | ((eptr[2] & 0x3f) << 12) | \
664              ((eptr[3] & 0x3f) << 6) | (eptr[4] & 0x3f); \
665          eptr += 5; \
666        } \        } \
667      }      }
668    
669    #if defined COMPILE_PCRE8
670    
671    /* These macros were originally written in the form of loops that used data
672    from the tables whose names start with PRIV(utf8_table). They were rewritten by
673    a user so as not to use loops, because in some environments this gives a
674    significant performance advantage, and it seems never to do any harm. */
675    
676    /* Tells the biggest code point which can be encoded as a single character. */
677    
678    #define MAX_VALUE_FOR_SINGLE_CHAR 127
679    
680    /* Tests whether the code point needs extra characters to decode. */
681    
682    #define HAS_EXTRALEN(c) ((c) >= 0xc0)
683    
684    /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
685    Otherwise it has an undefined behaviour. */
686    
687    #define GET_EXTRALEN(c) (PRIV(utf8_table4)[(c) & 0x3f])
688    
689    /* Returns TRUE, if the given character is not the first character
690    of a UTF sequence. */
691    
692    #define NOT_FIRSTCHAR(c) (((c) & 0xc0) == 0x80)
693    
694    /* Get the next UTF-8 character, not advancing the pointer. This is called when
695    we know we are in UTF-8 mode. */
696    
697    #define GETCHAR(c, eptr) \
698      c = *eptr; \
699      if (c >= 0xc0) GETUTF8(c, eptr);
700    
701  /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the  /* Get the next UTF-8 character, testing for UTF-8 mode, and not advancing the
702  pointer. */  pointer. */
703    
704  #define GETCHARTEST(c, eptr) \  #define GETCHARTEST(c, eptr) \
705    c = *eptr; \    c = *eptr; \
706    if (utf8 && c >= 0xc0) \    if (utf && c >= 0xc0) GETUTF8(c, eptr);
     { \  
     int gcii; \  
     int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  
     int gcss = 6*gcaa; \  
     c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
     for (gcii = 1; gcii <= gcaa; gcii++) \  
       { \  
       gcss -= 6; \  
       c |= (eptr[gcii] & 0x3f) << gcss; \  
       } \  
     }  
707    
708  /* Get the next UTF-8 character, advancing the pointer. This is called when we  /* Get the next UTF-8 character, advancing the pointer. This is called when we
709  know we are in UTF-8 mode. */  know we are in UTF-8 mode. */
710    
711  #define GETCHARINC(c, eptr) \  #define GETCHARINC(c, eptr) \
712    c = *eptr++; \    c = *eptr++; \
713    if (c >= 0xc0) \    if (c >= 0xc0) GETUTF8INC(c, eptr);
     { \  
     int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  
     int gcss = 6*gcaa; \  
     c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
     while (gcaa-- > 0) \  
       { \  
       gcss -= 6; \  
       c |= (*eptr++ & 0x3f) << gcss; \  
       } \  
     }  
714    
715  /* Get the next character, testing for UTF-8 mode, and advancing the pointer */  /* Get the next character, testing for UTF-8 mode, and advancing the pointer.
716    This is called when we don't know if we are in UTF-8 mode. */
717    
718  #define GETCHARINCTEST(c, eptr) \  #define GETCHARINCTEST(c, eptr) \
719    c = *eptr++; \    c = *eptr++; \
720    if (utf8 && c >= 0xc0) \    if (utf && c >= 0xc0) GETUTF8INC(c, eptr);
721    
722    /* Base macro to pick up the remaining bytes of a UTF-8 character, not
723    advancing the pointer, incrementing the length. */
724    
725    #define GETUTF8LEN(c, eptr, len) \
726      { \      { \
727      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \      if ((c & 0x20) == 0) \
     int gcss = 6*gcaa; \  
     c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
     while (gcaa-- > 0) \  
728        { \        { \
729        gcss -= 6; \        c = ((c & 0x1f) << 6) | (eptr[1] & 0x3f); \
730        c |= (*eptr++ & 0x3f) << gcss; \        len++; \
731          } \
732        else if ((c & 0x10)  == 0) \
733          { \
734          c = ((c & 0x0f) << 12) | ((eptr[1] & 0x3f) << 6) | (eptr[2] & 0x3f); \
735          len += 2; \
736          } \
737        else if ((c & 0x08)  == 0) \
738          {\
739          c = ((c & 0x07) << 18) | ((eptr[1] & 0x3f) << 12) | \
740              ((eptr[2] & 0x3f) << 6) | (eptr[3] & 0x3f); \
741          len += 3; \
742          } \
743        else if ((c & 0x04)  == 0) \
744          { \
745          c = ((c & 0x03) << 24) | ((eptr[1] & 0x3f) << 18) | \
746              ((eptr[2] & 0x3f) << 12) | ((eptr[3] & 0x3f) << 6) | \
747              (eptr[4] & 0x3f); \
748          len += 4; \
749          } \
750        else \
751          {\
752          c = ((c & 0x01) << 30) | ((eptr[1] & 0x3f) << 24) | \
753              ((eptr[2] & 0x3f) << 18) | ((eptr[3] & 0x3f) << 12) | \
754              ((eptr[4] & 0x3f) << 6) | (eptr[5] & 0x3f); \
755          len += 5; \
756        } \        } \
757      }      }
758    
# Line 446  if there are extra bytes. This is called Line 761  if there are extra bytes. This is called
761    
762  #define GETCHARLEN(c, eptr, len) \  #define GETCHARLEN(c, eptr, len) \
763    c = *eptr; \    c = *eptr; \
764    if (c >= 0xc0) \    if (c >= 0xc0) GETUTF8LEN(c, eptr, len);
765      { \  
766      int gcii; \  /* Get the next UTF-8 character, testing for UTF-8 mode, not advancing the
767      int gcaa = _pcre_utf8_table4[c & 0x3f];  /* Number of additional bytes */ \  pointer, incrementing length if there are extra bytes. This is called when we
768      int gcss = 6*gcaa; \  do not know if we are in UTF-8 mode. */
769      c = (c & _pcre_utf8_table3[gcaa]) << gcss; \  
770      for (gcii = 1; gcii <= gcaa; gcii++) \  #define GETCHARLENTEST(c, eptr, len) \
771        { \    c = *eptr; \
772        gcss -= 6; \    if (utf && c >= 0xc0) GETUTF8LEN(c, eptr, len);
       c |= (eptr[gcii] & 0x3f) << gcss; \  
       } \  
     len += gcaa; \  
     }  
773    
774  /* If the pointer is not at the start of a character, move it back until  /* If the pointer is not at the start of a character, move it back until
775  it is. This is called only in UTF-8 mode - we don't put a test within the macro  it is. This is called only in UTF-8 mode - we don't put a test within the macro
# Line 466  because almost all calls are already wit Line 777  because almost all calls are already wit
777    
778  #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--  #define BACKCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr--
779    
780  #endif  /* Same as above, just in the other direction. */
781    #define FORWARDCHAR(eptr) while((*eptr & 0xc0) == 0x80) eptr++
782    
783    /* Same as above, but it allows a fully customizable form. */
784    #define ACROSSCHAR(condition, eptr, action) \
785      while((condition) && ((eptr) & 0xc0) == 0x80) action
786    
787    #elif defined COMPILE_PCRE16
788    
789    /* Tells the biggest code point which can be encoded as a single character. */
790    
791  /* In case there is no definition of offsetof() provided - though any proper  #define MAX_VALUE_FOR_SINGLE_CHAR 65535
 Standard C system should have one. */  
792    
793  #ifndef offsetof  /* Tests whether the code point needs extra characters to decode. */
794  #define offsetof(p_type,field) ((size_t)&(((p_type *)0)->field))  
795    #define HAS_EXTRALEN(c) (((c) & 0xfc00) == 0xd800)
796    
797    /* Returns with the additional number of characters if IS_MULTICHAR(c) is TRUE.
798    Otherwise it has an undefined behaviour. */
799    
800    #define GET_EXTRALEN(c) 1
801    
802    /* Returns TRUE, if the given character is not the first character
803    of a UTF sequence. */
804    
805    #define NOT_FIRSTCHAR(c) (((c) & 0xfc00) == 0xdc00)
806    
807    /* Base macro to pick up the low surrogate of a UTF-16 character, not
808    advancing the pointer. */
809    
810    #define GETUTF16(c, eptr) \
811       { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; }
812    
813    /* Get the next UTF-16 character, not advancing the pointer. This is called when
814    we know we are in UTF-16 mode. */
815    
816    #define GETCHAR(c, eptr) \
817      c = *eptr; \
818      if ((c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
819    
820    /* Get the next UTF-16 character, testing for UTF-16 mode, and not advancing the
821    pointer. */
822    
823    #define GETCHARTEST(c, eptr) \
824      c = *eptr; \
825      if (utf && (c & 0xfc00) == 0xd800) GETUTF16(c, eptr);
826    
827    /* Base macro to pick up the low surrogate of a UTF-16 character, advancing
828    the pointer. */
829    
830    #define GETUTF16INC(c, eptr) \
831       { c = (((c & 0x3ff) << 10) | (*eptr++ & 0x3ff)) + 0x10000; }
832    
833    /* Get the next UTF-16 character, advancing the pointer. This is called when we
834    know we are in UTF-16 mode. */
835    
836    #define GETCHARINC(c, eptr) \
837      c = *eptr++; \
838      if ((c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
839    
840    /* Get the next character, testing for UTF-16 mode, and advancing the pointer.
841    This is called when we don't know if we are in UTF-16 mode. */
842    
843    #define GETCHARINCTEST(c, eptr) \
844      c = *eptr++; \
845      if (utf && (c & 0xfc00) == 0xd800) GETUTF16INC(c, eptr);
846    
847    /* Base macro to pick up the low surrogate of a UTF-16 character, not
848    advancing the pointer, incrementing the length. */
849    
850    #define GETUTF16LEN(c, eptr, len) \
851       { c = (((c & 0x3ff) << 10) | (eptr[1] & 0x3ff)) + 0x10000; len++; }
852    
853    /* Get the next UTF-16 character, not advancing the pointer, incrementing
854    length if there is a low surrogate. This is called when we know we are in
855    UTF-16 mode. */
856    
857    #define GETCHARLEN(c, eptr, len) \
858      c = *eptr; \
859      if ((c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
860    
861    /* Get the next UTF-816character, testing for UTF-16 mode, not advancing the
862    pointer, incrementing length if there is a low surrogate. This is called when
863    we do not know if we are in UTF-16 mode. */
864    
865    #define GETCHARLENTEST(c, eptr, len) \
866      c = *eptr; \
867      if (utf && (c & 0xfc00) == 0xd800) GETUTF16LEN(c, eptr, len);
868    
869    /* If the pointer is not at the start of a character, move it back until
870    it is. This is called only in UTF-16 mode - we don't put a test within the
871    macro because almost all calls are already within a block of UTF-16 only
872    code. */
873    
874    #define BACKCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr--
875    
876    /* Same as above, just in the other direction. */
877    #define FORWARDCHAR(eptr) if ((*eptr & 0xfc00) == 0xdc00) eptr++
878    
879    /* Same as above, but it allows a fully customizable form. */
880    #define ACROSSCHAR(condition, eptr, action) \
881      if ((condition) && ((eptr) & 0xfc00) == 0xdc00) action
882    
883    #elif defined COMPILE_PCRE32
884    
885    /* These are trivial for the 32-bit library, since all UTF-32 characters fit
886    into one pcre_uchar unit. */
887    #define MAX_VALUE_FOR_SINGLE_CHAR (0x10ffffu)
888    #define HAS_EXTRALEN(c) (0)
889    #define GET_EXTRALEN(c) (0)
890    #define NOT_FIRSTCHAR(c) (0)
891    
892    /* Get the next UTF-32 character, not advancing the pointer. This is called when
893    we know we are in UTF-32 mode. */
894    
895    #define GETCHAR(c, eptr) \
896      c = *(eptr);
897    
898    /* Get the next UTF-32 character, testing for UTF-32 mode, and not advancing the
899    pointer. */
900    
901    #define GETCHARTEST(c, eptr) \
902      c = *(eptr);
903    
904    /* Get the next UTF-32 character, advancing the pointer. This is called when we
905    know we are in UTF-32 mode. */
906    
907    #define GETCHARINC(c, eptr) \
908      c = *((eptr)++);
909    
910    /* Get the next character, testing for UTF-32 mode, and advancing the pointer.
911    This is called when we don't know if we are in UTF-32 mode. */
912    
913    #define GETCHARINCTEST(c, eptr) \
914      c = *((eptr)++);
915    
916    /* Get the next UTF-32 character, not advancing the pointer, not incrementing
917    length (since all UTF-32 is of length 1). This is called when we know we are in
918    UTF-32 mode. */
919    
920    #define GETCHARLEN(c, eptr, len) \
921      GETCHAR(c, eptr)
922    
923    /* Get the next UTF-32character, testing for UTF-32 mode, not advancing the
924    pointer, not incrementing the length (since all UTF-32 is of length 1).
925    This is called when we do not know if we are in UTF-32 mode. */
926    
927    #define GETCHARLENTEST(c, eptr, len) \
928      GETCHARTEST(c, eptr)
929    
930    /* If the pointer is not at the start of a character, move it back until
931    it is. This is called only in UTF-32 mode - we don't put a test within the
932    macro because almost all calls are already within a block of UTF-32 only
933    code.
934    These are all no-ops since all UTF-32 characters fit into one pcre_uchar. */
935    
936    #define BACKCHAR(eptr) do { } while (0)
937    
938    /* Same as above, just in the other direction. */
939    #define FORWARDCHAR(eptr) do { } while (0)
940    
941    /* Same as above, but it allows a fully customizable form. */
942    #define ACROSSCHAR(condition, eptr, action) do { } while (0)
943    
944    #else
945    #error Unsupported compiling mode
946    #endif /* COMPILE_PCRE[8|16|32] */
947    
948    #endif  /* SUPPORT_UTF */
949    
950    /* Tests for Unicode horizontal and vertical whitespace characters must check a
951    number of different values. Using a switch statement for this generates the
952    fastest code (no loop, no memory access), and there are several places in the
953    interpreter code where this happens. In order to ensure that all the case lists
954    remain in step, we use macros so that there is only one place where the lists
955    are defined.
956    
957    These values are also required as lists in pcre_compile.c when processing \h,
958    \H, \v and \V in a character class. The lists are defined in pcre_tables.c, but
959    macros that define the values are here so that all the definitions are
960    together. The lists must be in ascending character order, terminated by
961    NOTACHAR (which is 0xffffffff).
962    
963    Any changes should ensure that the various macros are kept in step with each
964    other. NOTE: The values also appear in pcre_jit_compile.c. */
965    
966    /* ------ ASCII/Unicode environments ------ */
967    
968    #ifndef EBCDIC
969    
970    #define HSPACE_LIST \
971      CHAR_HT, CHAR_SPACE, 0xa0, \
972      0x1680, 0x180e, 0x2000, 0x2001, 0x2002, 0x2003, 0x2004, 0x2005, \
973      0x2006, 0x2007, 0x2008, 0x2009, 0x200A, 0x202f, 0x205f, 0x3000, \
974      NOTACHAR
975    
976    #define HSPACE_MULTIBYTE_CASES \
977      case 0x1680:  /* OGHAM SPACE MARK */ \
978      case 0x180e:  /* MONGOLIAN VOWEL SEPARATOR */ \
979      case 0x2000:  /* EN QUAD */ \
980      case 0x2001:  /* EM QUAD */ \
981      case 0x2002:  /* EN SPACE */ \
982      case 0x2003:  /* EM SPACE */ \
983      case 0x2004:  /* THREE-PER-EM SPACE */ \
984      case 0x2005:  /* FOUR-PER-EM SPACE */ \
985      case 0x2006:  /* SIX-PER-EM SPACE */ \
986      case 0x2007:  /* FIGURE SPACE */ \
987      case 0x2008:  /* PUNCTUATION SPACE */ \
988      case 0x2009:  /* THIN SPACE */ \
989      case 0x200A:  /* HAIR SPACE */ \
990      case 0x202f:  /* NARROW NO-BREAK SPACE */ \
991      case 0x205f:  /* MEDIUM MATHEMATICAL SPACE */ \
992      case 0x3000   /* IDEOGRAPHIC SPACE */
993    
994    #define HSPACE_BYTE_CASES \
995      case CHAR_HT: \
996      case CHAR_SPACE: \
997      case 0xa0     /* NBSP */
998    
999    #define HSPACE_CASES \
1000      HSPACE_BYTE_CASES: \
1001      HSPACE_MULTIBYTE_CASES
1002    
1003    #define VSPACE_LIST \
1004      CHAR_LF, CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, 0x2028, 0x2029, NOTACHAR
1005    
1006    #define VSPACE_MULTIBYTE_CASES \
1007      case 0x2028:    /* LINE SEPARATOR */ \
1008      case 0x2029     /* PARAGRAPH SEPARATOR */
1009    
1010    #define VSPACE_BYTE_CASES \
1011      case CHAR_LF: \
1012      case CHAR_VT: \
1013      case CHAR_FF: \
1014      case CHAR_CR: \
1015      case CHAR_NEL
1016    
1017    #define VSPACE_CASES \
1018      VSPACE_BYTE_CASES: \
1019      VSPACE_MULTIBYTE_CASES
1020    
1021    /* ------ EBCDIC environments ------ */
1022    
1023    #else
1024    #define HSPACE_LIST CHAR_HT, CHAR_SPACE
1025    
1026    #define HSPACE_BYTE_CASES \
1027      case CHAR_HT: \
1028      case CHAR_SPACE
1029    
1030    #define HSPACE_CASES HSPACE_BYTE_CASES
1031    
1032    #ifdef EBCDIC_NL25
1033    #define VSPACE_LIST \
1034      CHAR_VT, CHAR_FF, CHAR_CR, CHAR_NEL, CHAR_LF, NOTACHAR
1035    #else
1036    #define VSPACE_LIST \
1037      CHAR_VT, CHAR_FF, CHAR_CR, CHAR_LF, CHAR_NEL, NOTACHAR
1038  #endif  #endif
1039    
1040    #define VSPACE_BYTE_CASES \
1041      case CHAR_LF: \
1042      case CHAR_VT: \
1043      case CHAR_FF: \
1044      case CHAR_CR: \
1045      case CHAR_NEL
1046    
1047  /* These are the public options that can change during matching. */  #define VSPACE_CASES VSPACE_BYTE_CASES
1048    #endif  /* EBCDIC */
1049    
1050  #define PCRE_IMS (PCRE_CASELESS|PCRE_MULTILINE|PCRE_DOTALL)  /* ------ End of whitespace macros ------ */
1051    
 /* Private flags containing information about the compiled regex. They used to  
 live at the top end of the options word, but that got almost full, so now they  
 are in a 16-bit flags word. */  
1052    
 #define PCRE_NOPARTIAL     0x0001  /* can't use partial with this regex */  
 #define PCRE_FIRSTSET      0x0002  /* first_byte is set */  
 #define PCRE_REQCHSET      0x0004  /* req_byte is set */  
 #define PCRE_STARTLINE     0x0008  /* start after \n for multiline */  
 #define PCRE_JCHANGED      0x0010  /* j option used in regex */  
 #define PCRE_HASCRORLF     0x0020  /* explicit \r or \n in pattern */  
1053    
1054  /* Options for the "extra" block produced by pcre_study(). */  /* Private flags containing information about the compiled regex. They used to
1055    live at the top end of the options word, but that got almost full, so they were
1056    moved to a 16-bit flags word - which got almost full, so now they are in a
1057    32-bit flags word. From release 8.00, PCRE_NOPARTIAL is unused, as the
1058    restrictions on partial matching have been lifted. It remains for backwards
1059    compatibility. */
1060    
1061    #define PCRE_MODE8         0x00000001  /* compiled in 8 bit mode */
1062    #define PCRE_MODE16        0x00000002  /* compiled in 16 bit mode */
1063    #define PCRE_MODE32        0x00000004  /* compiled in 32 bit mode */
1064    #define PCRE_FIRSTSET      0x00000010  /* first_char is set */
1065    #define PCRE_FCH_CASELESS  0x00000020  /* caseless first char */
1066    #define PCRE_REQCHSET      0x00000040  /* req_byte is set */
1067    #define PCRE_RCH_CASELESS  0x00000080  /* caseless requested char */
1068    #define PCRE_STARTLINE     0x00000100  /* start after \n for multiline */
1069    #define PCRE_NOPARTIAL     0x00000200  /* can't use partial with this regex */
1070    #define PCRE_JCHANGED      0x00000400  /* j option used in regex */
1071    #define PCRE_HASCRORLF     0x00000800  /* explicit \r or \n in pattern */
1072    #define PCRE_HASTHEN       0x00001000  /* pattern contains (*THEN) */
1073    #define PCRE_MLSET         0x00002000  /* match limit set by regex */
1074    #define PCRE_RLSET         0x00004000  /* recursion limit set by regex */
1075    #define PCRE_MATCH_EMPTY   0x00008000  /* pattern can match empty string */
1076    
1077    #if defined COMPILE_PCRE8
1078    #define PCRE_MODE          PCRE_MODE8
1079    #elif defined COMPILE_PCRE16
1080    #define PCRE_MODE          PCRE_MODE16
1081    #elif defined COMPILE_PCRE32
1082    #define PCRE_MODE          PCRE_MODE32
1083    #endif
1084    #define PCRE_MODE_MASK     (PCRE_MODE8 | PCRE_MODE16 | PCRE_MODE32)
1085    
1086    /* Flags for the "extra" block produced by pcre_study(). */
1087    
1088  #define PCRE_STUDY_MAPPED   0x01     /* a map of starting chars exists */  #define PCRE_STUDY_MAPPED  0x0001  /* a map of starting chars exists */
1089    #define PCRE_STUDY_MINLEN  0x0002  /* a minimum length field exists */
1090    
1091  /* Masks for identifying the public options that are permitted at compile  /* Masks for identifying the public options that are permitted at compile
1092  time, run time, or study time, respectively. */  time, run time, or study time, respectively. */
# Line 502  time, run time, or study time, respectiv Line 1094  time, run time, or study time, respectiv
1094  #define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \  #define PCRE_NEWLINE_BITS (PCRE_NEWLINE_CR|PCRE_NEWLINE_LF|PCRE_NEWLINE_ANY| \
1095                             PCRE_NEWLINE_ANYCRLF)                             PCRE_NEWLINE_ANYCRLF)
1096    
1097  #define PUBLIC_OPTIONS \  #define PUBLIC_COMPILE_OPTIONS \
1098    (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \    (PCRE_CASELESS|PCRE_EXTENDED|PCRE_ANCHORED|PCRE_MULTILINE| \
1099     PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \     PCRE_DOTALL|PCRE_DOLLAR_ENDONLY|PCRE_EXTRA|PCRE_UNGREEDY|PCRE_UTF8| \
1100     PCRE_NO_AUTO_CAPTURE|PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \     PCRE_NO_AUTO_CAPTURE|PCRE_NO_AUTO_POSSESS| \
1101     PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)     PCRE_NO_UTF8_CHECK|PCRE_AUTO_CALLOUT|PCRE_FIRSTLINE| \
1102       PCRE_DUPNAMES|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
1103       PCRE_JAVASCRIPT_COMPAT|PCRE_UCP|PCRE_NO_START_OPTIMIZE|PCRE_NEVER_UTF)
1104    
1105  #define PUBLIC_EXEC_OPTIONS \  #define PUBLIC_EXEC_OPTIONS \
1106    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
1107     PCRE_PARTIAL|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)     PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_NEWLINE_BITS| \
1108       PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE|PCRE_NO_START_OPTIMIZE)
1109    
1110  #define PUBLIC_DFA_EXEC_OPTIONS \  #define PUBLIC_DFA_EXEC_OPTIONS \
1111    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NO_UTF8_CHECK| \    (PCRE_ANCHORED|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|PCRE_NOTEMPTY_ATSTART| \
1112     PCRE_PARTIAL|PCRE_DFA_SHORTEST|PCRE_DFA_RESTART|PCRE_NEWLINE_BITS| \     PCRE_NO_UTF8_CHECK|PCRE_PARTIAL_HARD|PCRE_PARTIAL_SOFT|PCRE_DFA_SHORTEST| \
1113     PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE)     PCRE_DFA_RESTART|PCRE_NEWLINE_BITS|PCRE_BSR_ANYCRLF|PCRE_BSR_UNICODE| \
1114       PCRE_NO_START_OPTIMIZE)
1115  #define PUBLIC_STUDY_OPTIONS 0   /* None defined */  
1116    #define PUBLIC_STUDY_OPTIONS \
1117       (PCRE_STUDY_JIT_COMPILE|PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE| \
1118        PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE|PCRE_STUDY_EXTRA_NEEDED)
1119    
1120    #define PUBLIC_JIT_EXEC_OPTIONS \
1121       (PCRE_NO_UTF8_CHECK|PCRE_NOTBOL|PCRE_NOTEOL|PCRE_NOTEMPTY|\
1122        PCRE_NOTEMPTY_ATSTART|PCRE_PARTIAL_SOFT|PCRE_PARTIAL_HARD)
1123    
1124  /* Magic number to provide a small check against being handed junk. Also used  /* Magic number to provide a small check against being handed junk. */
 to detect whether a pattern was compiled on a host of different endianness. */  
1125    
1126  #define MAGIC_NUMBER  0x50435245UL   /* 'PCRE' */  #define MAGIC_NUMBER  0x50435245UL   /* 'PCRE' */
1127    
1128  /* Negative values for the firstchar and reqchar variables */  /* This variable is used to detect a loaded regular expression
1129    in different endianness. */
1130    
1131  #define REQ_UNSET (-2)  #define REVERSED_MAGIC_NUMBER  0x45524350UL   /* 'ERCP' */
 #define REQ_NONE  (-1)  
1132    
1133  /* The maximum remaining length of subject we are prepared to search for a  /* The maximum remaining length of subject we are prepared to search for a
1134  req_byte match. */  req_byte match. */
1135    
1136  #define REQ_BYTE_MAX 1000  #define REQ_BYTE_MAX 1000
1137    
1138  /* Flags added to firstbyte or reqbyte; a "non-literal" item is either a  /* Miscellaneous definitions. The #ifndef is to pacify compiler warnings in
1139  variable-length repeat, or a anything other than literal characters. */  environments where these macros are defined elsewhere. Unfortunately, there
1140    is no way to do the same for the typedef. */
 #define REQ_CASELESS 0x0100    /* indicates caselessness */  
 #define REQ_VARY     0x0200    /* reqbyte followed non-literal item */  
   
 /* Miscellaneous definitions */  
1141    
1142  typedef int BOOL;  typedef int BOOL;
1143    
1144    #ifndef FALSE
1145  #define FALSE   0  #define FALSE   0
1146  #define TRUE    1  #define TRUE    1
1147    #endif
1148    
1149    /* If PCRE is to support UTF-8 on EBCDIC platforms, we cannot use normal
1150    character constants like '*' because the compiler would emit their EBCDIC code,
1151    which is different from their ASCII/UTF-8 code. Instead we define macros for
1152    the characters so that they always use the ASCII/UTF-8 code when UTF-8 support
1153    is enabled. When UTF-8 support is not enabled, the definitions use character
1154    literals. Both character and string versions of each character are needed, and
1155    there are some longer strings as well.
1156    
1157    This means that, on EBCDIC platforms, the PCRE library can handle either
1158    EBCDIC, or UTF-8, but not both. To support both in the same compiled library
1159    would need different lookups depending on whether PCRE_UTF8 was set or not.
1160    This would make it impossible to use characters in switch/case statements,
1161    which would reduce performance. For a theoretical use (which nobody has asked
1162    for) in a minority area (EBCDIC platforms), this is not sensible. Any
1163    application that did need both could compile two versions of the library, using
1164    macros to give the functions distinct names. */
1165    
1166    #ifndef SUPPORT_UTF
1167    
1168    /* UTF-8 support is not enabled; use the platform-dependent character literals
1169    so that PCRE works in both ASCII and EBCDIC environments, but only in non-UTF
1170    mode. Newline characters are problematic in EBCDIC. Though it has CR and LF
1171    characters, a common practice has been to use its NL (0x15) character as the
1172    line terminator in C-like processing environments. However, sometimes the LF
1173    (0x25) character is used instead, according to this Unicode document:
1174    
1175    http://unicode.org/standard/reports/tr13/tr13-5.html
1176    
1177    PCRE defaults EBCDIC NL to 0x15, but has a build-time option to select 0x25
1178    instead. Whichever is *not* chosen is defined as NEL.
1179    
1180    In both ASCII and EBCDIC environments, CHAR_NL and CHAR_LF are synonyms for the
1181    same code point. */
1182    
1183    #ifdef EBCDIC
1184    
1185    #ifndef EBCDIC_NL25
1186    #define CHAR_NL                     '\x15'
1187    #define CHAR_NEL                    '\x25'
1188    #define STR_NL                      "\x15"
1189    #define STR_NEL                     "\x25"
1190    #else
1191    #define CHAR_NL                     '\x25'
1192    #define CHAR_NEL                    '\x15'
1193    #define STR_NL                      "\x25"
1194    #define STR_NEL                     "\x15"
1195    #endif
1196    
1197    #define CHAR_LF                     CHAR_NL
1198    #define STR_LF                      STR_NL
1199    
1200    #define CHAR_ESC                    '\047'
1201    #define CHAR_DEL                    '\007'
1202    #define STR_ESC                     "\047"
1203    #define STR_DEL                     "\007"
1204    
1205    #else  /* Not EBCDIC */
1206    
1207    /* In ASCII/Unicode, linefeed is '\n' and we equate this to NL for
1208    compatibility. NEL is the Unicode newline character; make sure it is
1209    a positive value. */
1210    
1211    #define CHAR_LF                     '\n'
1212    #define CHAR_NL                     CHAR_LF
1213    #define CHAR_NEL                    ((unsigned char)'\x85')
1214    #define CHAR_ESC                    '\033'
1215    #define CHAR_DEL                    '\177'
1216    
1217    #define STR_LF                      "\n"
1218    #define STR_NL                      STR_LF
1219    #define STR_NEL                     "\x85"
1220    #define STR_ESC                     "\033"
1221    #define STR_DEL                     "\177"
1222    
1223    #endif  /* EBCDIC */
1224    
1225    /* The remaining definitions work in both environments. */
1226    
1227    #define CHAR_NULL                   '\0'
1228    #define CHAR_HT                     '\t'
1229    #define CHAR_VT                     '\v'
1230    #define CHAR_FF                     '\f'
1231    #define CHAR_CR                     '\r'
1232    #define CHAR_BS                     '\b'
1233    #define CHAR_BEL                    '\a'
1234    
1235    #define CHAR_SPACE                  ' '
1236    #define CHAR_EXCLAMATION_MARK       '!'
1237    #define CHAR_QUOTATION_MARK         '"'
1238    #define CHAR_NUMBER_SIGN            '#'
1239    #define CHAR_DOLLAR_SIGN            '$'
1240    #define CHAR_PERCENT_SIGN           '%'
1241    #define CHAR_AMPERSAND              '&'
1242    #define CHAR_APOSTROPHE             '\''
1243    #define CHAR_LEFT_PARENTHESIS       '('
1244    #define CHAR_RIGHT_PARENTHESIS      ')'
1245    #define CHAR_ASTERISK               '*'
1246    #define CHAR_PLUS                   '+'
1247    #define CHAR_COMMA                  ','
1248    #define CHAR_MINUS                  '-'
1249    #define CHAR_DOT                    '.'
1250    #define CHAR_SLASH                  '/'
1251    #define CHAR_0                      '0'
1252    #define CHAR_1                      '1'
1253    #define CHAR_2                      '2'
1254    #define CHAR_3                      '3'
1255    #define CHAR_4                      '4'
1256    #define CHAR_5                      '5'
1257    #define CHAR_6                      '6'
1258    #define CHAR_7                      '7'
1259    #define CHAR_8                      '8'
1260    #define CHAR_9                      '9'
1261    #define CHAR_COLON                  ':'
1262    #define CHAR_SEMICOLON              ';'
1263    #define CHAR_LESS_THAN_SIGN         '<'
1264    #define CHAR_EQUALS_SIGN            '='
1265    #define CHAR_GREATER_THAN_SIGN      '>'
1266    #define CHAR_QUESTION_MARK          '?'
1267    #define CHAR_COMMERCIAL_AT          '@'
1268    #define CHAR_A                      'A'
1269    #define CHAR_B                      'B'
1270    #define CHAR_C                      'C'
1271    #define CHAR_D                      'D'
1272    #define CHAR_E                      'E'
1273    #define CHAR_F                      'F'
1274    #define CHAR_G                      'G'
1275    #define CHAR_H                      'H'
1276    #define CHAR_I                      'I'
1277    #define CHAR_J                      'J'
1278    #define CHAR_K                      'K'
1279    #define CHAR_L                      'L'
1280    #define CHAR_M                      'M'
1281    #define CHAR_N                      'N'
1282    #define CHAR_O                      'O'
1283    #define CHAR_P                      'P'
1284    #define CHAR_Q                      'Q'
1285    #define CHAR_R                      'R'
1286    #define CHAR_S                      'S'
1287    #define CHAR_T                      'T'
1288    #define CHAR_U                      'U'
1289    #define CHAR_V                      'V'
1290    #define CHAR_W                      'W'
1291    #define CHAR_X                      'X'
1292    #define CHAR_Y                      'Y'
1293    #define CHAR_Z                      'Z'
1294    #define CHAR_LEFT_SQUARE_BRACKET    '['
1295    #define CHAR_BACKSLASH              '\\'
1296    #define CHAR_RIGHT_SQUARE_BRACKET   ']'
1297    #define CHAR_CIRCUMFLEX_ACCENT      '^'
1298    #define CHAR_UNDERSCORE             '_'
1299    #define CHAR_GRAVE_ACCENT           '`'
1300    #define CHAR_a                      'a'
1301    #define CHAR_b                      'b'
1302    #define CHAR_c                      'c'
1303    #define CHAR_d                      'd'
1304    #define CHAR_e                      'e'
1305    #define CHAR_f                      'f'
1306    #define CHAR_g                      'g'
1307    #define CHAR_h                      'h'
1308    #define CHAR_i                      'i'
1309    #define CHAR_j                      'j'
1310    #define CHAR_k                      'k'
1311    #define CHAR_l                      'l'
1312    #define CHAR_m                      'm'
1313    #define CHAR_n                      'n'
1314    #define CHAR_o                      'o'
1315    #define CHAR_p                      'p'
1316    #define CHAR_q                      'q'
1317    #define CHAR_r                      'r'
1318    #define CHAR_s                      's'
1319    #define CHAR_t                      't'
1320    #define CHAR_u                      'u'
1321    #define CHAR_v                      'v'
1322    #define CHAR_w                      'w'
1323    #define CHAR_x                      'x'
1324    #define CHAR_y                      'y'
1325    #define CHAR_z                      'z'
1326    #define CHAR_LEFT_CURLY_BRACKET     '{'
1327    #define CHAR_VERTICAL_LINE          '|'
1328    #define CHAR_RIGHT_CURLY_BRACKET    '}'
1329    #define CHAR_TILDE                  '~'
1330    
1331    #define STR_HT                      "\t"
1332    #define STR_VT                      "\v"
1333    #define STR_FF                      "\f"
1334    #define STR_CR                      "\r"
1335    #define STR_BS                      "\b"
1336    #define STR_BEL                     "\a"
1337    
1338    #define STR_SPACE                   " "
1339    #define STR_EXCLAMATION_MARK        "!"
1340    #define STR_QUOTATION_MARK          "\""
1341    #define STR_NUMBER_SIGN             "#"
1342    #define STR_DOLLAR_SIGN             "$"
1343    #define STR_PERCENT_SIGN            "%"
1344    #define STR_AMPERSAND               "&"
1345    #define STR_APOSTROPHE              "'"
1346    #define STR_LEFT_PARENTHESIS        "("
1347    #define STR_RIGHT_PARENTHESIS       ")"
1348    #define STR_ASTERISK                "*"
1349    #define STR_PLUS                    "+"
1350    #define STR_COMMA                   ","
1351    #define STR_MINUS                   "-"
1352    #define STR_DOT                     "."
1353    #define STR_SLASH                   "/"
1354    #define STR_0                       "0"
1355    #define STR_1                       "1"
1356    #define STR_2                       "2"
1357    #define STR_3                       "3"
1358    #define STR_4                       "4"
1359    #define STR_5                       "5"
1360    #define STR_6                       "6"
1361    #define STR_7                       "7"
1362    #define STR_8                       "8"
1363    #define STR_9                       "9"
1364    #define STR_COLON                   ":"
1365    #define STR_SEMICOLON               ";"
1366    #define STR_LESS_THAN_SIGN          "<"
1367    #define STR_EQUALS_SIGN             "="
1368    #define STR_GREATER_THAN_SIGN       ">"
1369    #define STR_QUESTION_MARK           "?"
1370    #define STR_COMMERCIAL_AT           "@"
1371    #define STR_A                       "A"
1372    #define STR_B                       "B"
1373    #define STR_C                       "C"
1374    #define STR_D                       "D"
1375    #define STR_E                       "E"
1376    #define STR_F                       "F"
1377    #define STR_G                       "G"
1378    #define STR_H                       "H"
1379    #define STR_I                       "I"
1380    #define STR_J                       "J"
1381    #define STR_K                       "K"
1382    #define STR_L                       "L"
1383    #define STR_M                       "M"
1384    #define STR_N                       "N"
1385    #define STR_O                       "O"
1386    #define STR_P                       "P"
1387    #define STR_Q                       "Q"
1388    #define STR_R                       "R"
1389    #define STR_S                       "S"
1390    #define STR_T                       "T"
1391    #define STR_U                       "U"
1392    #define STR_V                       "V"
1393    #define STR_W                       "W"
1394    #define STR_X                       "X"
1395    #define STR_Y                       "Y"
1396    #define STR_Z                       "Z"
1397    #define STR_LEFT_SQUARE_BRACKET     "["
1398    #define STR_BACKSLASH               "\\"
1399    #define STR_RIGHT_SQUARE_BRACKET    "]"
1400    #define STR_CIRCUMFLEX_ACCENT       "^"
1401    #define STR_UNDERSCORE              "_"
1402    #define STR_GRAVE_ACCENT            "`"
1403    #define STR_a                       "a"
1404    #define STR_b                       "b"
1405    #define STR_c                       "c"
1406    #define STR_d                       "d"
1407    #define STR_e                       "e"
1408    #define STR_f                       "f"
1409    #define STR_g                       "g"
1410    #define STR_h                       "h"
1411    #define STR_i                       "i"
1412    #define STR_j                       "j"
1413    #define STR_k                       "k"
1414    #define STR_l                       "l"
1415    #define STR_m                       "m"
1416    #define STR_n                       "n"
1417    #define STR_o                       "o"
1418    #define STR_p                       "p"
1419    #define STR_q                       "q"
1420    #define STR_r                       "r"
1421    #define STR_s                       "s"
1422    #define STR_t                       "t"
1423    #define STR_u                       "u"
1424    #define STR_v                       "v"
1425    #define STR_w                       "w"
1426    #define STR_x                       "x"
1427    #define STR_y                       "y"
1428    #define STR_z                       "z"
1429    #define STR_LEFT_CURLY_BRACKET      "{"
1430    #define STR_VERTICAL_LINE           "|"
1431    #define STR_RIGHT_CURLY_BRACKET     "}"
1432    #define STR_TILDE                   "~"
1433    
1434    #define STRING_ACCEPT0              "ACCEPT\0"
1435    #define STRING_COMMIT0              "COMMIT\0"
1436    #define STRING_F0                   "F\0"
1437    #define STRING_FAIL0                "FAIL\0"
1438    #define STRING_MARK0                "MARK\0"
1439    #define STRING_PRUNE0               "PRUNE\0"
1440    #define STRING_SKIP0                "SKIP\0"
1441    #define STRING_THEN                 "THEN"
1442    
1443    #define STRING_alpha0               "alpha\0"
1444    #define STRING_lower0               "lower\0"
1445    #define STRING_upper0               "upper\0"
1446    #define STRING_alnum0               "alnum\0"
1447    #define STRING_ascii0               "ascii\0"
1448    #define STRING_blank0               "blank\0"
1449    #define STRING_cntrl0               "cntrl\0"
1450    #define STRING_digit0               "digit\0"
1451    #define STRING_graph0               "graph\0"
1452    #define STRING_print0               "print\0"
1453    #define STRING_punct0               "punct\0"
1454    #define STRING_space0               "space\0"
1455    #define STRING_word0                "word\0"
1456    #define STRING_xdigit               "xdigit"
1457    
1458    #define STRING_DEFINE               "DEFINE"
1459    #define STRING_WEIRD_STARTWORD      "[:<:]]"
1460    #define STRING_WEIRD_ENDWORD        "[:>:]]"
1461    
1462    #define STRING_CR_RIGHTPAR              "CR)"
1463    #define STRING_LF_RIGHTPAR              "LF)"
1464    #define STRING_CRLF_RIGHTPAR            "CRLF)"
1465    #define STRING_ANY_RIGHTPAR             "ANY)"
1466    #define STRING_ANYCRLF_RIGHTPAR         "ANYCRLF)"
1467    #define STRING_BSR_ANYCRLF_RIGHTPAR     "BSR_ANYCRLF)"
1468    #define STRING_BSR_UNICODE_RIGHTPAR     "BSR_UNICODE)"
1469    #define STRING_UTF8_RIGHTPAR            "UTF8)"
1470    #define STRING_UTF16_RIGHTPAR           "UTF16)"
1471    #define STRING_UTF32_RIGHTPAR           "UTF32)"
1472    #define STRING_UTF_RIGHTPAR             "UTF)"
1473    #define STRING_UCP_RIGHTPAR             "UCP)"
1474    #define STRING_NO_AUTO_POSSESS_RIGHTPAR "NO_AUTO_POSSESS)"
1475    #define STRING_NO_START_OPT_RIGHTPAR    "NO_START_OPT)"
1476    #define STRING_LIMIT_MATCH_EQ           "LIMIT_MATCH="
1477    #define STRING_LIMIT_RECURSION_EQ       "LIMIT_RECURSION="
1478    
1479    #else  /* SUPPORT_UTF */
1480    
1481    /* UTF-8 support is enabled; always use UTF-8 (=ASCII) character codes. This
1482    works in both modes non-EBCDIC platforms, and on EBCDIC platforms in UTF-8 mode
1483    only. */
1484    
1485    #define CHAR_HT                     '\011'
1486    #define CHAR_VT                     '\013'
1487    #define CHAR_FF                     '\014'
1488    #define CHAR_CR                     '\015'
1489    #define CHAR_LF                     '\012'
1490    #define CHAR_NL                     CHAR_LF
1491    #define CHAR_NEL                    ((unsigned char)'\x85')
1492    #define CHAR_BS                     '\010'
1493    #define CHAR_BEL                    '\007'
1494    #define CHAR_ESC                    '\033'
1495    #define CHAR_DEL                    '\177'
1496    
1497    #define CHAR_NULL                   '\0'
1498    #define CHAR_SPACE                  '\040'
1499    #define CHAR_EXCLAMATION_MARK       '\041'
1500    #define CHAR_QUOTATION_MARK         '\042'
1501    #define CHAR_NUMBER_SIGN            '\043'
1502    #define CHAR_DOLLAR_SIGN            '\044'
1503    #define CHAR_PERCENT_SIGN           '\045'
1504    #define CHAR_AMPERSAND              '\046'
1505    #define CHAR_APOSTROPHE             '\047'
1506    #define CHAR_LEFT_PARENTHESIS       '\050'
1507    #define CHAR_RIGHT_PARENTHESIS      '\051'
1508    #define CHAR_ASTERISK               '\052'
1509    #define CHAR_PLUS                   '\053'
1510    #define CHAR_COMMA                  '\054'
1511    #define CHAR_MINUS                  '\055'
1512    #define CHAR_DOT                    '\056'
1513    #define CHAR_SLASH                  '\057'
1514    #define CHAR_0                      '\060'
1515    #define CHAR_1                      '\061'
1516    #define CHAR_2                      '\062'
1517    #define CHAR_3                      '\063'
1518    #define CHAR_4                      '\064'
1519    #define CHAR_5                      '\065'
1520    #define CHAR_6                      '\066'
1521    #define CHAR_7                      '\067'
1522    #define CHAR_8                      '\070'
1523    #define CHAR_9                      '\071'
1524    #define CHAR_COLON                  '\072'
1525    #define CHAR_SEMICOLON              '\073'
1526    #define CHAR_LESS_THAN_SIGN         '\074'
1527    #define CHAR_EQUALS_SIGN            '\075'
1528    #define CHAR_GREATER_THAN_SIGN      '\076'
1529    #define CHAR_QUESTION_MARK          '\077'
1530    #define CHAR_COMMERCIAL_AT          '\100'
1531    #define CHAR_A                      '\101'
1532    #define CHAR_B                      '\102'
1533    #define CHAR_C                      '\103'
1534    #define CHAR_D                      '\104'
1535    #define CHAR_E                      '\105'
1536    #define CHAR_F                      '\106'
1537    #define CHAR_G                      '\107'
1538    #define CHAR_H                      '\110'
1539    #define CHAR_I                      '\111'
1540    #define CHAR_J                      '\112'
1541    #define CHAR_K                      '\113'
1542    #define CHAR_L                      '\114'
1543    #define CHAR_M                      '\115'
1544    #define CHAR_N                      '\116'
1545    #define CHAR_O                      '\117'
1546    #define CHAR_P                      '\120'
1547    #define CHAR_Q                      '\121'
1548    #define CHAR_R                      '\122'
1549    #define CHAR_S                      '\123'
1550    #define CHAR_T                      '\124'
1551    #define CHAR_U                      '\125'
1552    #define CHAR_V                      '\126'
1553    #define CHAR_W                      '\127'
1554    #define CHAR_X                      '\130'
1555    #define CHAR_Y                      '\131'
1556    #define CHAR_Z                      '\132'
1557    #define CHAR_LEFT_SQUARE_BRACKET    '\133'
1558    #define CHAR_BACKSLASH              '\134'
1559    #define CHAR_RIGHT_SQUARE_BRACKET   '\135'
1560    #define CHAR_CIRCUMFLEX_ACCENT      '\136'
1561    #define CHAR_UNDERSCORE             '\137'
1562    #define CHAR_GRAVE_ACCENT           '\140'
1563    #define CHAR_a                      '\141'
1564    #define CHAR_b                      '\142'
1565    #define CHAR_c                      '\143'
1566    #define CHAR_d                      '\144'
1567    #define CHAR_e                      '\145'
1568    #define CHAR_f                      '\146'
1569    #define CHAR_g                      '\147'
1570    #define CHAR_h                      '\150'
1571    #define CHAR_i                      '\151'
1572    #define CHAR_j                      '\152'
1573    #define CHAR_k                      '\153'
1574    #define CHAR_l                      '\154'
1575    #define CHAR_m                      '\155'
1576    #define CHAR_n                      '\156'
1577    #define CHAR_o                      '\157'
1578    #define CHAR_p                      '\160'
1579    #define CHAR_q                      '\161'
1580    #define CHAR_r                      '\162'
1581    #define CHAR_s                      '\163'
1582    #define CHAR_t                      '\164'
1583    #define CHAR_u                      '\165'
1584    #define CHAR_v                      '\166'
1585    #define CHAR_w                      '\167'
1586    #define CHAR_x                      '\170'
1587    #define CHAR_y                      '\171'
1588    #define CHAR_z                      '\172'
1589    #define CHAR_LEFT_CURLY_BRACKET     '\173'
1590    #define CHAR_VERTICAL_LINE          '\174'
1591    #define CHAR_RIGHT_CURLY_BRACKET    '\175'
1592    #define CHAR_TILDE                  '\176'
1593    
1594    #define STR_HT                      "\011"
1595    #define STR_VT                      "\013"
1596    #define STR_FF                      "\014"
1597    #define STR_CR                      "\015"
1598    #define STR_NL                      "\012"
1599    #define STR_BS                      "\010"
1600    #define STR_BEL                     "\007"
1601    #define STR_ESC                     "\033"
1602    #define STR_DEL                     "\177"
1603    
1604    #define STR_SPACE                   "\040"
1605    #define STR_EXCLAMATION_MARK        "\041"
1606    #define STR_QUOTATION_MARK          "\042"
1607    #define STR_NUMBER_SIGN             "\043"
1608    #define STR_DOLLAR_SIGN             "\044"
1609    #define STR_PERCENT_SIGN            "\045"
1610    #define STR_AMPERSAND               "\046"
1611    #define STR_APOSTROPHE              "\047"
1612    #define STR_LEFT_PARENTHESIS        "\050"
1613    #define STR_RIGHT_PARENTHESIS       "\051"
1614    #define STR_ASTERISK                "\052"
1615    #define STR_PLUS                    "\053"
1616    #define STR_COMMA                   "\054"
1617    #define STR_MINUS                   "\055"
1618    #define STR_DOT                     "\056"
1619    #define STR_SLASH                   "\057"
1620    #define STR_0                       "\060"
1621    #define STR_1                       "\061"
1622    #define STR_2                       "\062"
1623    #define STR_3                       "\063"
1624    #define STR_4                       "\064"
1625    #define STR_5                       "\065"
1626    #define STR_6                       "\066"
1627    #define STR_7                       "\067"
1628    #define STR_8                       "\070"
1629    #define STR_9                       "\071"
1630    #define STR_COLON                   "\072"
1631    #define STR_SEMICOLON               "\073"
1632    #define STR_LESS_THAN_SIGN          "\074"
1633    #define STR_EQUALS_SIGN             "\075"
1634    #define STR_GREATER_THAN_SIGN       "\076"
1635    #define STR_QUESTION_MARK           "\077"
1636    #define STR_COMMERCIAL_AT           "\100"
1637    #define STR_A                       "\101"
1638    #define STR_B                       "\102"
1639    #define STR_C                       "\103"
1640    #define STR_D                       "\104"
1641    #define STR_E                       "\105"
1642    #define STR_F                       "\106"
1643    #define STR_G                       "\107"
1644    #define STR_H                       "\110"
1645    #define STR_I                       "\111"
1646    #define STR_J                       "\112"
1647    #define STR_K                       "\113"
1648    #define STR_L                       "\114"
1649    #define STR_M                       "\115"
1650    #define STR_N                       "\116"
1651    #define STR_O                       "\117"
1652    #define STR_P                       "\120"
1653    #define STR_Q                       "\121"
1654    #define STR_R                       "\122"
1655    #define STR_S                       "\123"
1656    #define STR_T                       "\124"
1657    #define STR_U                       "\125"
1658    #define STR_V                       "\126"
1659    #define STR_W                       "\127"
1660    #define STR_X                       "\130"
1661    #define STR_Y                       "\131"
1662    #define STR_Z                       "\132"
1663    #define STR_LEFT_SQUARE_BRACKET     "\133"
1664    #define STR_BACKSLASH               "\134"
1665    #define STR_RIGHT_SQUARE_BRACKET    "\135"
1666    #define STR_CIRCUMFLEX_ACCENT       "\136"
1667    #define STR_UNDERSCORE              "\137"
1668    #define STR_GRAVE_ACCENT            "\140"
1669    #define STR_a                       "\141"
1670    #define STR_b                       "\142"
1671    #define STR_c                       "\143"
1672    #define STR_d                       "\144"
1673    #define STR_e                       "\145"
1674    #define STR_f                       "\146"
1675    #define STR_g                       "\147"
1676    #define STR_h                       "\150"
1677    #define STR_i                       "\151"
1678    #define STR_j                       "\152"
1679    #define STR_k                       "\153"
1680    #define STR_l                       "\154"
1681    #define STR_m                       "\155"
1682    #define STR_n                       "\156"
1683    #define STR_o                       "\157"
1684    #define STR_p                       "\160"
1685    #define STR_q                       "\161"
1686    #define STR_r                       "\162"
1687    #define STR_s                       "\163"
1688    #define STR_t                       "\164"
1689    #define STR_u                       "\165"
1690    #define STR_v                       "\166"
1691    #define STR_w                       "\167"
1692    #define STR_x                       "\170"
1693    #define STR_y                       "\171"
1694    #define STR_z                       "\172"
1695    #define STR_LEFT_CURLY_BRACKET      "\173"
1696    #define STR_VERTICAL_LINE           "\174"
1697    #define STR_RIGHT_CURLY_BRACKET     "\175"
1698    #define STR_TILDE                   "\176"
1699    
1700    #define STRING_ACCEPT0              STR_A STR_C STR_C STR_E STR_P STR_T "\0"
1701    #define STRING_COMMIT0              STR_C STR_O STR_M STR_M STR_I STR_T "\0"
1702    #define STRING_F0                   STR_F "\0"
1703    #define STRING_FAIL0                STR_F STR_A STR_I STR_L "\0"
1704    #define STRING_MARK0                STR_M STR_A STR_R STR_K "\0"
1705    #define STRING_PRUNE0               STR_P STR_R STR_U STR_N STR_E "\0"
1706    #define STRING_SKIP0                STR_S STR_K STR_I STR_P "\0"
1707    #define STRING_THEN                 STR_T STR_H STR_E STR_N
1708    
1709    #define STRING_alpha0               STR_a STR_l STR_p STR_h STR_a "\0"
1710    #define STRING_lower0               STR_l STR_o STR_w STR_e STR_r "\0"
1711    #define STRING_upper0               STR_u STR_p STR_p STR_e STR_r "\0"
1712    #define STRING_alnum0               STR_a STR_l STR_n STR_u STR_m "\0"
1713    #define STRING_ascii0               STR_a STR_s STR_c STR_i STR_i "\0"
1714    #define STRING_blank0               STR_b STR_l STR_a STR_n STR_k "\0"
1715    #define STRING_cntrl0               STR_c STR_n STR_t STR_r STR_l "\0"
1716    #define STRING_digit0               STR_d STR_i STR_g STR_i STR_t "\0"
1717    #define STRING_graph0               STR_g STR_r STR_a STR_p STR_h "\0"
1718    #define STRING_print0               STR_p STR_r STR_i STR_n STR_t "\0"
1719    #define STRING_punct0               STR_p STR_u STR_n STR_c STR_t "\0"
1720    #define STRING_space0               STR_s STR_p STR_a STR_c STR_e "\0"
1721    #define STRING_word0                STR_w STR_o STR_r STR_d       "\0"
1722    #define STRING_xdigit               STR_x STR_d STR_i STR_g STR_i STR_t
1723    
1724    #define STRING_DEFINE               STR_D STR_E STR_F STR_I STR_N STR_E
1725    #define STRING_WEIRD_STARTWORD      STR_LEFT_SQUARE_BRACKET STR_COLON STR_LESS_THAN_SIGN STR_COLON STR_RIGHT_SQUARE_BRACKET STR_RIGHT_SQUARE_BRACKET
1726    #define STRING_WEIRD_ENDWORD        STR_LEFT_SQUARE_BRACKET STR_COLON STR_GREATER_THAN_SIGN STR_COLON STR_RIGHT_SQUARE_BRACKET STR_RIGHT_SQUARE_BRACKET
1727    
1728    #define STRING_CR_RIGHTPAR              STR_C STR_R STR_RIGHT_PARENTHESIS
1729    #define STRING_LF_RIGHTPAR              STR_L STR_F STR_RIGHT_PARENTHESIS
1730    #define STRING_CRLF_RIGHTPAR            STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1731    #define STRING_ANY_RIGHTPAR             STR_A STR_N STR_Y STR_RIGHT_PARENTHESIS
1732    #define STRING_ANYCRLF_RIGHTPAR         STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1733    #define STRING_BSR_ANYCRLF_RIGHTPAR     STR_B STR_S STR_R STR_UNDERSCORE STR_A STR_N STR_Y STR_C STR_R STR_L STR_F STR_RIGHT_PARENTHESIS
1734    #define STRING_BSR_UNICODE_RIGHTPAR     STR_B STR_S STR_R STR_UNDERSCORE STR_U STR_N STR_I STR_C STR_O STR_D STR_E STR_RIGHT_PARENTHESIS
1735    #define STRING_UTF8_RIGHTPAR            STR_U STR_T STR_F STR_8 STR_RIGHT_PARENTHESIS
1736    #define STRING_UTF16_RIGHTPAR           STR_U STR_T STR_F STR_1 STR_6 STR_RIGHT_PARENTHESIS
1737    #define STRING_UTF32_RIGHTPAR           STR_U STR_T STR_F STR_3 STR_2 STR_RIGHT_PARENTHESIS
1738    #define STRING_UTF_RIGHTPAR             STR_U STR_T STR_F STR_RIGHT_PARENTHESIS
1739    #define STRING_UCP_RIGHTPAR             STR_U STR_C STR_P STR_RIGHT_PARENTHESIS
1740    #define STRING_NO_AUTO_POSSESS_RIGHTPAR STR_N STR_O STR_UNDERSCORE STR_A STR_U STR_T STR_O STR_UNDERSCORE STR_P STR_O STR_S STR_S STR_E STR_S STR_S STR_RIGHT_PARENTHESIS
1741    #define STRING_NO_START_OPT_RIGHTPAR    STR_N STR_O STR_UNDERSCORE STR_S STR_T STR_A STR_R STR_T STR_UNDERSCORE STR_O STR_P STR_T STR_RIGHT_PARENTHESIS
1742    #define STRING_LIMIT_MATCH_EQ           STR_L STR_I STR_M STR_I STR_T STR_UNDERSCORE STR_M STR_A STR_T STR_C STR_H STR_EQUALS_SIGN
1743    #define STRING_LIMIT_RECURSION_EQ       STR_L STR_I STR_M STR_I STR_T STR_UNDERSCORE STR_R STR_E STR_C STR_U STR_R STR_S STR_I STR_O STR_N STR_EQUALS_SIGN
1744    
1745    #endif  /* SUPPORT_UTF */
1746    
1747  /* Escape items that are just an encoding of a particular data value. */  /* Escape items that are just an encoding of a particular data value. */
1748    
1749  #ifndef ESC_e  #ifndef ESC_e
1750  #define ESC_e 27  #define ESC_e CHAR_ESC
1751  #endif  #endif
1752    
1753  #ifndef ESC_f  #ifndef ESC_f
1754  #define ESC_f '\f'  #define ESC_f CHAR_FF
1755  #endif  #endif
1756    
1757  #ifndef ESC_n  #ifndef ESC_n
1758  #define ESC_n '\n'  #define ESC_n CHAR_LF
1759  #endif  #endif
1760    
1761  #ifndef ESC_r  #ifndef ESC_r
1762  #define ESC_r '\r'  #define ESC_r CHAR_CR
1763  #endif  #endif
1764    
1765  /* We can't officially use ESC_t because it is a POSIX reserved identifier  /* We can't officially use ESC_t because it is a POSIX reserved identifier
1766  (presumably because of all the others like size_t). */  (presumably because of all the others like size_t). */
1767    
1768  #ifndef ESC_tee  #ifndef ESC_tee
1769  #define ESC_tee '\t'  #define ESC_tee CHAR_HT
1770  #endif  #endif
1771    
1772  /* Codes for different types of Unicode property */  /* Codes for different types of Unicode property */
1773    
1774  #define PT_ANY        0    /* Any property - matches all chars */  #define PT_ANY        0    /* Any property - matches all chars */
1775  #define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */  #define PT_LAMP       1    /* L& - the union of Lu, Ll, Lt */
1776  #define PT_GC         2    /* General characteristic (e.g. L) */  #define PT_GC         2    /* Specified general characteristic (e.g. L) */
1777  #define PT_PC         3    /* Particular characteristic (e.g. Lu) */  #define PT_PC         3    /* Specified particular characteristic (e.g. Lu) */
1778  #define PT_SC         4    /* Script (e.g. Han) */  #define PT_SC         4    /* Script (e.g. Han) */
1779    #define PT_ALNUM      5    /* Alphanumeric - the union of L and N */
1780    #define PT_SPACE      6    /* Perl space - Z plus 9,10,12,13 */
1781    #define PT_PXSPACE    7    /* POSIX space - Z plus 9,10,11,12,13 */
1782    #define PT_WORD       8    /* Word - L plus N plus underscore */
1783    #define PT_CLIST      9    /* Pseudo-property: match character list */
1784    #define PT_UCNC      10    /* Universal Character nameable character */
1785    #define PT_TABSIZE   11    /* Size of square table for autopossessify tests */
1786    
1787    /* The following special properties are used only in XCLASS items, when POSIX
1788    classes are specified and PCRE_UCP is set - in other words, for Unicode
1789    handling of these classes. They are not available via the \p or \P escapes like
1790    those in the above list, and so they do not take part in the autopossessifying
1791    table. */
1792    
1793    #define PT_PXGRAPH   11    /* [:graph:] - characters that mark the paper */
1794    #define PT_PXPRINT   12    /* [:print:] - [:graph:] plus non-control spaces */
1795    #define PT_PXPUNCT   13    /* [:punct:] - punctuation characters */
1796    
1797  /* Flag bits and data types for the extended class (OP_XCLASS) for classes that  /* Flag bits and data types for the extended class (OP_XCLASS) for classes that
1798  contain UTF-8 characters with values greater than 255. */  contain characters with values greater than 255. */
1799    
1800  #define XCL_NOT    0x01    /* Flag: this is a negative class */  #define XCL_NOT       0x01    /* Flag: this is a negative class */
1801  #define XCL_MAP    0x02    /* Flag: a 32-byte map is present */  #define XCL_MAP       0x02    /* Flag: a 32-byte map is present */
1802    #define XCL_HASPROP   0x04    /* Flag: property checks are present. */
1803    
1804  #define XCL_END       0    /* Marks end of individual items */  #define XCL_END       0    /* Marks end of individual items */
1805  #define XCL_SINGLE    1    /* Single item (one multibyte char) follows */  #define XCL_SINGLE    1    /* Single item (one multibyte char) follows */
# Line 593  contain UTF-8 characters with values gre Line 1808  contain UTF-8 characters with values gre
1808  #define XCL_NOTPROP   4    /* Unicode inverted property (ditto) */  #define XCL_NOTPROP   4    /* Unicode inverted property (ditto) */
1809    
1810  /* These are escaped items that aren't just an encoding of a particular data  /* These are escaped items that aren't just an encoding of a particular data
1811  value such as \n. They must have non-zero values, as check_escape() returns  value such as \n. They must have non-zero values, as check_escape() returns 0
1812  their negation. Also, they must appear in the same order as in the opcode  for a data character.  Also, they must appear in the same order as in the
1813  definitions below, up to ESC_z. There's a dummy for OP_ANY because it  opcode definitions below, up to ESC_z. There's a dummy for OP_ALLANY because it
1814  corresponds to "." rather than an escape sequence. The final one must be  corresponds to "." in DOTALL mode rather than an escape sequence. It is also
1815  ESC_REF as subsequent values are used for backreferences (\1, \2, \3, etc).  used for [^] in JavaScript compatibility mode, and for \C in non-utf mode. In
1816  There are two tests in the code for an escape greater than ESC_b and less than  non-DOTALL mode, "." behaves like \N.
1817  ESC_Z to detect the types that may be repeated. These are the types that  
1818  consume characters. If any new escapes are put in between that don't consume a  The special values ESC_DU, ESC_du, etc. are used instead of ESC_D, ESC_d, etc.
1819  character, that code will have to change. */  when PCRE_UCP is set and replacement of \d etc by \p sequences is required.
1820    They must be contiguous, and remain in order so that the replacements can be
1821    looked up from a table.
1822    
1823    Negative numbers are used to encode a backreference (\1, \2, \3, etc.) in
1824    check_escape(). There are two tests in the code for an escape
1825    greater than ESC_b and less than ESC_Z to detect the types that may be
1826    repeated. These are the types that consume characters. If any new escapes are
1827    put in between that don't consume a character, that code will have to change.
1828    */
1829    
1830  enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,  enum { ESC_A = 1, ESC_G, ESC_K, ESC_B, ESC_b, ESC_D, ESC_d, ESC_S, ESC_s,
1831         ESC_W, ESC_w, ESC_dum1, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H, ESC_h,         ESC_W, ESC_w, ESC_N, ESC_dum, ESC_C, ESC_P, ESC_p, ESC_R, ESC_H,
1832         ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z, ESC_E, ESC_Q, ESC_k, ESC_REF };         ESC_h, ESC_V, ESC_v, ESC_X, ESC_Z, ESC_z,
1833           ESC_E, ESC_Q, ESC_g, ESC_k,
1834           ESC_DU, ESC_du, ESC_SU, ESC_su, ESC_WU, ESC_wu };
1835    
1836    
1837    /********************** Opcode definitions ******************/
1838    
1839    /****** NOTE NOTE NOTE ******
1840    
1841    Starting from 1 (i.e. after OP_END), the values up to OP_EOD must correspond in
1842    order to the list of escapes immediately above. Furthermore, values up to
1843    OP_DOLLM must not be changed without adjusting the table called autoposstab in
1844    pcre_compile.c
1845    
1846    Whenever this list is updated, the two macro definitions that follow must be
1847    updated to match. The possessification table called "opcode_possessify" in
1848    pcre_compile.c must also be updated, and also the tables called "coptable"
1849    and "poptable" in pcre_dfa_exec.c.
1850    
1851    ****** NOTE NOTE NOTE ******/
1852    
 /* Opcode table: Starting from 1 (i.e. after OP_END), the values up to  
 OP_EOD must correspond in order to the list of escapes immediately above.  
1853    
1854  *** NOTE NOTE NOTE *** Whenever this list is updated, the two macro definitions  /* The values between FIRST_AUTOTAB_OP and LAST_AUTOTAB_RIGHT_OP, inclusive,
1855  that follow must also be updated to match. There is also a table called  are used in a table for deciding whether a repeated character type can be
1856  "coptable" in pcre_dfa_exec.c that must be updated. */  auto-possessified. */
1857    
1858    #define FIRST_AUTOTAB_OP       OP_NOT_DIGIT
1859    #define LAST_AUTOTAB_LEFT_OP   OP_EXTUNI
1860    #define LAST_AUTOTAB_RIGHT_OP  OP_DOLLM
1861    
1862  enum {  enum {
1863    OP_END,            /* 0 End of pattern */    OP_END,            /* 0 End of pattern */
# Line 631  enum { Line 1875  enum {
1875    OP_WHITESPACE,         /*  9 \s */    OP_WHITESPACE,         /*  9 \s */
1876    OP_NOT_WORDCHAR,       /* 10 \W */    OP_NOT_WORDCHAR,       /* 10 \W */
1877    OP_WORDCHAR,           /* 11 \w */    OP_WORDCHAR,           /* 11 \w */
   OP_ANY,            /* 12 Match any character */  
   OP_ANYBYTE,        /* 13 Match any byte (\C); different to OP_ANY for UTF-8 */  
   OP_NOTPROP,        /* 14 \P (not Unicode property) */  
   OP_PROP,           /* 15 \p (Unicode property) */  
   OP_ANYNL,          /* 16 \R (any newline sequence) */  
   OP_NOT_HSPACE,     /* 17 \H (not horizontal whitespace) */  
   OP_HSPACE,         /* 18 \h (horizontal whitespace) */  
   OP_NOT_VSPACE,     /* 19 \V (not vertical whitespace) */  
   OP_VSPACE,         /* 20 \v (vertical whitespace) */  
   OP_EXTUNI,         /* 21 \X (extended Unicode sequence */  
   OP_EODN,           /* 22 End of data or \n at end of data: \Z. */  
   OP_EOD,            /* 23 End of data: \z */  
   
   OP_OPT,            /* 24 Set runtime options */  
   OP_CIRC,           /* 25 Start of line - varies with multiline switch */  
   OP_DOLL,           /* 26 End of line - varies with multiline switch */  
   OP_CHAR,           /* 27 Match one character, casefully */  
   OP_CHARNC,         /* 28 Match one character, caselessly */  
   OP_NOT,            /* 29 Match one character, not the following one */  
   
   OP_STAR,           /* 30 The maximizing and minimizing versions of */  
   OP_MINSTAR,        /* 31 these six opcodes must come in pairs, with */  
   OP_PLUS,           /* 32 the minimizing one second. */  
   OP_MINPLUS,        /* 33 This first set applies to single characters.*/  
   OP_QUERY,          /* 34 */  
   OP_MINQUERY,       /* 35 */  
   
   OP_UPTO,           /* 36 From 0 to n matches */  
   OP_MINUPTO,        /* 37 */  
   OP_EXACT,          /* 38 Exactly n matches */  
   
   OP_POSSTAR,        /* 39 Possessified star */  
   OP_POSPLUS,        /* 40 Possessified plus */  
   OP_POSQUERY,       /* 41 Posesssified query */  
   OP_POSUPTO,        /* 42 Possessified upto */  
   
   OP_NOTSTAR,        /* 43 The maximizing and minimizing versions of */  
   OP_NOTMINSTAR,     /* 44 these six opcodes must come in pairs, with */  
   OP_NOTPLUS,        /* 45 the minimizing one second. They must be in */  
   OP_NOTMINPLUS,     /* 46 exactly the same order as those above. */  
   OP_NOTQUERY,       /* 47 This set applies to "not" single characters. */  
   OP_NOTMINQUERY,    /* 48 */  
   
   OP_NOTUPTO,        /* 49 From 0 to n matches */  
   OP_NOTMINUPTO,     /* 50 */  
   OP_NOTEXACT,       /* 51 Exactly n matches */  
   
   OP_NOTPOSSTAR,     /* 52 Possessified versions */  
   OP_NOTPOSPLUS,     /* 53 */  
   OP_NOTPOSQUERY,    /* 54 */  
   OP_NOTPOSUPTO,     /* 55 */  
   
   OP_TYPESTAR,       /* 56 The maximizing and minimizing versions of */  
   OP_TYPEMINSTAR,    /* 57 these six opcodes must come in pairs, with */  
   OP_TYPEPLUS,       /* 58 the minimizing one second. These codes must */  
   OP_TYPEMINPLUS,    /* 59 be in exactly the same order as those above. */  
   OP_TYPEQUERY,      /* 60 This set applies to character types such as \d */  
   OP_TYPEMINQUERY,   /* 61 */  
   
   OP_TYPEUPTO,       /* 62 From 0 to n matches */  
   OP_TYPEMINUPTO,    /* 63 */  
   OP_TYPEEXACT,      /* 64 Exactly n matches */  
   
   OP_TYPEPOSSTAR,    /* 65 Possessified versions */  
   OP_TYPEPOSPLUS,    /* 66 */  
   OP_TYPEPOSQUERY,   /* 67 */  
   OP_TYPEPOSUPTO,    /* 68 */  
   
   OP_CRSTAR,         /* 69 The maximizing and minimizing versions of */  
   OP_CRMINSTAR,      /* 70 all these opcodes must come in pairs, with */  
   OP_CRPLUS,         /* 71 the minimizing one second. These codes must */  
   OP_CRMINPLUS,      /* 72 be in exactly the same order as those above. */  
   OP_CRQUERY,        /* 73 These are for character classes and back refs */  
   OP_CRMINQUERY,     /* 74 */  
   OP_CRRANGE,        /* 75 These are different to the three sets above. */  
   OP_CRMINRANGE,     /* 76 */  
   
   OP_CLASS,          /* 77 Match a character class, chars < 256 only */  
   OP_NCLASS,         /* 78 Same, but the bitmap was created from a negative  
                            class - the difference is relevant only when a UTF-8  
                            character > 255 is encountered. */  
   
   OP_XCLASS,         /* 79 Extended class for handling UTF-8 chars within the  
                            class. This does both positive and negative. */  
   
   OP_REF,            /* 80 Match a back reference */  
   OP_RECURSE,        /* 81 Match a numbered subpattern (possibly recursive) */  
   OP_CALLOUT,        /* 82 Call out to external function if provided */  
   
   OP_ALT,            /* 83 Start of alternation */  
   OP_KET,            /* 84 End of group that doesn't have an unbounded repeat */  
   OP_KETRMAX,        /* 85 These two must remain together and in this */  
   OP_KETRMIN,        /* 86 order. They are for groups the repeat for ever. */  
   
   /* The assertions must come before BRA, CBRA, ONCE, and COND.*/  
   
   OP_ASSERT,         /* 87 Positive lookahead */  
   OP_ASSERT_NOT,     /* 88 Negative lookahead */  
   OP_ASSERTBACK,     /* 89 Positive lookbehind */  
   OP_ASSERTBACK_NOT, /* 90 Negative lookbehind */  
   OP_REVERSE,        /* 91 Move pointer back - used in lookbehind assertions */  
   
   /* ONCE, BRA, CBRA, and COND must come after the assertions, with ONCE first,  
   as there's a test for >= ONCE for a subpattern that isn't an assertion. */  
   
   OP_ONCE,           /* 92 Atomic group */  
   OP_BRA,            /* 93 Start of non-capturing bracket */  
   OP_CBRA,           /* 94 Start of capturing bracket */  
   OP_COND,           /* 95 Conditional group */  
1878    
1879    /* These three must follow the previous three, in the same order. There's a    OP_ANY,            /* 12 Match any character except newline (\N) */
1880    check for >= SBRA to distinguish the two sets. */    OP_ALLANY,         /* 13 Match any character */
1881      OP_ANYBYTE,        /* 14 Match any byte (\C); different to OP_ANY for UTF-8 */
1882      OP_NOTPROP,        /* 15 \P (not Unicode property) */
1883      OP_PROP,           /* 16 \p (Unicode property) */
1884      OP_ANYNL,          /* 17 \R (any newline sequence) */
1885      OP_NOT_HSPACE,     /* 18 \H (not horizontal whitespace) */
1886      OP_HSPACE,         /* 19 \h (horizontal whitespace) */
1887      OP_NOT_VSPACE,     /* 20 \V (not vertical whitespace) */
1888      OP_VSPACE,         /* 21 \v (vertical whitespace) */
1889      OP_EXTUNI,         /* 22 \X (extended Unicode sequence */
1890      OP_EODN,           /* 23 End of data or \n at end of data (\Z) */
1891      OP_EOD,            /* 24 End of data (\z) */
1892    
1893      /* Line end assertions */
1894    
1895      OP_DOLL,           /* 25 End of line - not multiline */
1896      OP_DOLLM,          /* 26 End of line - multiline */
1897      OP_CIRC,           /* 27 Start of line - not multiline */
1898      OP_CIRCM,          /* 28 Start of line - multiline */
1899    
1900      /* Single characters; caseful must precede the caseless ones */
1901    
1902      OP_CHAR,           /* 29 Match one character, casefully */
1903      OP_CHARI,          /* 30 Match one character, caselessly */
1904      OP_NOT,            /* 31 Match one character, not the given one, casefully */
1905      OP_NOTI,           /* 32 Match one character, not the given one, caselessly */
1906    
1907      /* The following sets of 13 opcodes must always be kept in step because
1908      the offset from the first one is used to generate the others. */
1909    
1910      /* Repeated characters; caseful must precede the caseless ones */
1911    
1912      OP_STAR,           /* 33 The maximizing and minimizing versions of */
1913      OP_MINSTAR,        /* 34 these six opcodes must come in pairs, with */
1914      OP_PLUS,           /* 35 the minimizing one second. */
1915      OP_MINPLUS,        /* 36 */
1916      OP_QUERY,          /* 37 */
1917      OP_MINQUERY,       /* 38 */
1918    
1919      OP_UPTO,           /* 39 From 0 to n matches of one character, caseful*/
1920      OP_MINUPTO,        /* 40 */
1921      OP_EXACT,          /* 41 Exactly n matches */
1922    
1923      OP_POSSTAR,        /* 42 Possessified star, caseful */
1924      OP_POSPLUS,        /* 43 Possessified plus, caseful */
1925      OP_POSQUERY,       /* 44 Posesssified query, caseful */
1926      OP_POSUPTO,        /* 45 Possessified upto, caseful */
1927    
1928      /* Repeated characters; caseless must follow the caseful ones */
1929    
1930      OP_STARI,          /* 46 */
1931      OP_MINSTARI,       /* 47 */
1932      OP_PLUSI,          /* 48 */
1933      OP_MINPLUSI,       /* 49 */
1934      OP_QUERYI,         /* 50 */
1935      OP_MINQUERYI,      /* 51 */
1936    
1937      OP_UPTOI,          /* 52 From 0 to n matches of one character, caseless */
1938      OP_MINUPTOI,       /* 53 */
1939      OP_EXACTI,         /* 54 */
1940    
1941      OP_POSSTARI,       /* 55 Possessified star, caseless */
1942      OP_POSPLUSI,       /* 56 Possessified plus, caseless */
1943      OP_POSQUERYI,      /* 57 Posesssified query, caseless */
1944      OP_POSUPTOI,       /* 58 Possessified upto, caseless */
1945    
1946      /* The negated ones must follow the non-negated ones, and match them */
1947      /* Negated repeated character, caseful; must precede the caseless ones */
1948    
1949      OP_NOTSTAR,        /* 59 The maximizing and minimizing versions of */
1950      OP_NOTMINSTAR,     /* 60 these six opcodes must come in pairs, with */
1951      OP_NOTPLUS,        /* 61 the minimizing one second. They must be in */
1952      OP_NOTMINPLUS,     /* 62 exactly the same order as those above. */
1953      OP_NOTQUERY,       /* 63 */
1954      OP_NOTMINQUERY,    /* 64 */
1955    
1956      OP_NOTUPTO,        /* 65 From 0 to n matches, caseful */
1957      OP_NOTMINUPTO,     /* 66 */
1958      OP_NOTEXACT,       /* 67 Exactly n matches */
1959    
1960      OP_NOTPOSSTAR,     /* 68 Possessified versions, caseful */
1961      OP_NOTPOSPLUS,     /* 69 */
1962      OP_NOTPOSQUERY,    /* 70 */
1963      OP_NOTPOSUPTO,     /* 71 */
1964    
1965      /* Negated repeated character, caseless; must follow the caseful ones */
1966    
1967      OP_NOTSTARI,       /* 72 */
1968      OP_NOTMINSTARI,    /* 73 */
1969      OP_NOTPLUSI,       /* 74 */
1970      OP_NOTMINPLUSI,    /* 75 */
1971      OP_NOTQUERYI,      /* 76 */
1972      OP_NOTMINQUERYI,   /* 77 */
1973    
1974      OP_NOTUPTOI,       /* 78 From 0 to n matches, caseless */
1975      OP_NOTMINUPTOI,    /* 79 */
1976      OP_NOTEXACTI,      /* 80 Exactly n matches */
1977    
1978      OP_NOTPOSSTARI,    /* 81 Possessified versions, caseless */
1979      OP_NOTPOSPLUSI,    /* 82 */
1980      OP_NOTPOSQUERYI,   /* 83 */
1981      OP_NOTPOSUPTOI,    /* 84 */
1982    
1983      /* Character types */
1984    
1985      OP_TYPESTAR,       /* 85 The maximizing and minimizing versions of */
1986      OP_TYPEMINSTAR,    /* 86 these six opcodes must come in pairs, with */
1987      OP_TYPEPLUS,       /* 87 the minimizing one second. These codes must */
1988      OP_TYPEMINPLUS,    /* 88 be in exactly the same order as those above. */
1989      OP_TYPEQUERY,      /* 89 */
1990      OP_TYPEMINQUERY,   /* 90 */
1991    
1992      OP_TYPEUPTO,       /* 91 From 0 to n matches */
1993      OP_TYPEMINUPTO,    /* 92 */
1994      OP_TYPEEXACT,      /* 93 Exactly n matches */
1995    
1996      OP_TYPEPOSSTAR,    /* 94 Possessified versions */
1997      OP_TYPEPOSPLUS,    /* 95 */
1998      OP_TYPEPOSQUERY,   /* 96 */
1999      OP_TYPEPOSUPTO,    /* 97 */
2000    
2001      /* These are used for character classes and back references; only the
2002      first six are the same as the sets above. */
2003    
2004      OP_CRSTAR,         /* 98 The maximizing and minimizing versions of */
2005      OP_CRMINSTAR,      /* 99 all these opcodes must come in pairs, with */
2006      OP_CRPLUS,         /* 100 the minimizing one second. These codes must */
2007      OP_CRMINPLUS,      /* 101 be in exactly the same order as those above. */
2008      OP_CRQUERY,        /* 102 */
2009      OP_CRMINQUERY,     /* 103 */
2010    
2011      OP_CRRANGE,        /* 104 These are different to the three sets above. */
2012      OP_CRMINRANGE,     /* 105 */
2013    
2014      OP_CRPOSSTAR,      /* 106 Possessified versions */
2015      OP_CRPOSPLUS,      /* 107 */
2016      OP_CRPOSQUERY,     /* 108 */
2017      OP_CRPOSRANGE,     /* 109 */
2018    
2019      /* End of quantifier opcodes */
2020    
2021      OP_CLASS,          /* 110 Match a character class, chars < 256 only */
2022      OP_NCLASS,         /* 111 Same, but the bitmap was created from a negative
2023                                  class - the difference is relevant only when a
2024                                  character > 255 is encountered. */
2025      OP_XCLASS,         /* 112 Extended class for handling > 255 chars within the
2026                                  class. This does both positive and negative. */
2027      OP_REF,            /* 113 Match a back reference, casefully */
2028      OP_REFI,           /* 114 Match a back reference, caselessly */
2029      OP_DNREF,          /* 115 Match a duplicate name backref, casefully */
2030      OP_DNREFI,         /* 116 Match a duplicate name backref, caselessly */
2031      OP_RECURSE,        /* 117 Match a numbered subpattern (possibly recursive) */
2032      OP_CALLOUT,        /* 118 Call out to external function if provided */
2033    
2034      OP_ALT,            /* 119 Start of alternation */
2035      OP_KET,            /* 120 End of group that doesn't have an unbounded repeat */
2036      OP_KETRMAX,        /* 121 These two must remain together and in this */
2037      OP_KETRMIN,        /* 122 order. They are for groups the repeat for ever. */
2038      OP_KETRPOS,        /* 123 Possessive unlimited repeat. */
2039    
2040      /* The assertions must come before BRA, CBRA, ONCE, and COND, and the four
2041      asserts must remain in order. */
2042    
2043      OP_REVERSE,        /* 124 Move pointer back - used in lookbehind assertions */
2044      OP_ASSERT,         /* 125 Positive lookahead */
2045      OP_ASSERT_NOT,     /* 126 Negative lookahead */
2046      OP_ASSERTBACK,     /* 127 Positive lookbehind */
2047      OP_ASSERTBACK_NOT, /* 128 Negative lookbehind */
2048    
2049      /* ONCE, ONCE_NC, BRA, BRAPOS, CBRA, CBRAPOS, and COND must come immediately
2050      after the assertions, with ONCE first, as there's a test for >= ONCE for a
2051      subpattern that isn't an assertion. The POS versions must immediately follow
2052      the non-POS versions in each case. */
2053    
2054      OP_ONCE,           /* 129 Atomic group, contains captures */
2055      OP_ONCE_NC,        /* 130 Atomic group containing no captures */
2056      OP_BRA,            /* 131 Start of non-capturing bracket */
2057      OP_BRAPOS,         /* 132 Ditto, with unlimited, possessive repeat */
2058      OP_CBRA,           /* 133 Start of capturing bracket */
2059      OP_CBRAPOS,        /* 134 Ditto, with unlimited, possessive repeat */
2060      OP_COND,           /* 135 Conditional group */
2061    
2062    OP_SBRA,           /* 96 Start of non-capturing bracket, check empty  */    /* These five must follow the previous five, in the same order. There's a
2063    OP_SCBRA,          /* 97 Start of capturing bracket, check empty */    check for >= SBRA to distinguish the two sets. */
   OP_SCOND,          /* 98 Conditional group, check empty */  
   
   OP_CREF,           /* 99 Used to hold a capture number as condition */  
   OP_RREF,           /* 100 Used to hold a recursion number as condition */  
   OP_DEF,            /* 101 The DEFINE condition */  
2064    
2065    OP_BRAZERO,        /* 102 These two must remain together and in this */    OP_SBRA,           /* 136 Start of non-capturing bracket, check empty  */
2066    OP_BRAMINZERO,     /* 103 order. */    OP_SBRAPOS,        /* 137 Ditto, with unlimited, possessive repeat */
2067      OP_SCBRA,          /* 138 Start of capturing bracket, check empty */
2068      OP_SCBRAPOS,       /* 139 Ditto, with unlimited, possessive repeat */
2069      OP_SCOND,          /* 140 Conditional group, check empty */
2070    
2071      /* The next two pairs must (respectively) be kept together. */
2072    
2073      OP_CREF,           /* 141 Used to hold a capture number as condition */
2074      OP_DNCREF,         /* 142 Used to point to duplicate names as a condition */
2075      OP_RREF,           /* 143 Used to hold a recursion number as condition */
2076      OP_DNRREF,         /* 144 Used to point to duplicate names as a condition */
2077      OP_DEF,            /* 145 The DEFINE condition */
2078    
2079      OP_BRAZERO,        /* 146 These two must remain together and in this */
2080      OP_BRAMINZERO,     /* 147 order. */
2081      OP_BRAPOSZERO,     /* 148 */
2082    
2083    /* These are backtracking control verbs */    /* These are backtracking control verbs */
2084    
2085    OP_PRUNE,          /* 104 */    OP_MARK,           /* 149 always has an argument */
2086    OP_SKIP,           /* 105 */    OP_PRUNE,          /* 150 */
2087    OP_THEN,           /* 106 */    OP_PRUNE_ARG,      /* 151 same, but with argument */
2088    OP_COMMIT,         /* 107 */    OP_SKIP,           /* 152 */
2089      OP_SKIP_ARG,       /* 153 same, but with argument */
2090      OP_THEN,           /* 154 */
2091      OP_THEN_ARG,       /* 155 same, but with argument */
2092      OP_COMMIT,         /* 156 */
2093    
2094    /* These are forced failure and success verbs */    /* These are forced failure and success verbs */
2095    
2096    OP_FAIL,           /* 108 */    OP_FAIL,           /* 157 */
2097    OP_ACCEPT          /* 109 */    OP_ACCEPT,         /* 158 */
2098      OP_ASSERT_ACCEPT,  /* 159 Used inside assertions */
2099      OP_CLOSE,          /* 160 Used before OP_ACCEPT to close open captures */
2100    
2101      /* This is used to skip a subpattern with a {0} quantifier */
2102    
2103      OP_SKIPZERO,       /* 161 */
2104    
2105      /* This is not an opcode, but is used to check that tables indexed by opcode
2106      are the correct length, in order to catch updating errors - there have been
2107      some in the past. */
2108    
2109      OP_TABLE_LENGTH
2110  };  };
2111    
2112    /* *** NOTE NOTE NOTE *** Whenever the list above is updated, the two macro
2113    definitions that follow must also be updated to match. There are also tables
2114    called "opcode_possessify" in pcre_compile.c and "coptable" and "poptable" in
2115    pcre_dfa_exec.c that must be updated. */
2116    
2117    
2118  /* This macro defines textual names for all the opcodes. These are used only  /* This macro defines textual names for all the opcodes. These are used only
2119  for debugging. The macro is referenced only in pcre_printint.c. */  for debugging, and some of them are only partial names. The macro is referenced
2120    only in pcre_printint.c, which fills out the full names in many cases (and in
2121    some cases doesn't actually use these names at all). */
2122    
2123  #define OP_NAME_LIST \  #define OP_NAME_LIST \
2124    "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d",         \    "End", "\\A", "\\G", "\\K", "\\B", "\\b", "\\D", "\\d",         \
2125    "\\S", "\\s", "\\W", "\\w", "Any", "Anybyte",                   \    "\\S", "\\s", "\\W", "\\w", "Any", "AllAny", "Anybyte",         \
2126    "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v",           \    "notprop", "prop", "\\R", "\\H", "\\h", "\\V", "\\v",           \
2127    "extuni",  "\\Z", "\\z",                                        \    "extuni",  "\\Z", "\\z",                                        \
2128    "Opt", "^", "$", "char", "charnc", "not",                       \    "$", "$", "^", "^", "char", "chari", "not", "noti",             \
2129    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??",                                \
2130      "{", "{", "{",                                                  \
2131    "*+","++", "?+", "{",                                           \    "*+","++", "?+", "{",                                           \
2132    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??",                                \
2133      "{", "{", "{",                                                  \
2134      "*+","++", "?+", "{",                                           \
2135      "*", "*?", "+", "+?", "?", "??",                                \
2136      "{", "{", "{",                                                  \
2137      "*+","++", "?+", "{",                                           \
2138      "*", "*?", "+", "+?", "?", "??",                                \
2139      "{", "{", "{",                                                  \
2140    "*+","++", "?+", "{",                                           \    "*+","++", "?+", "{",                                           \
2141    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \    "*", "*?", "+", "+?", "?", "??", "{", "{", "{",                 \
2142    "*+","++", "?+", "{",                                           \    "*+","++", "?+", "{",                                           \
2143    "*", "*?", "+", "+?", "?", "??", "{", "{",                      \    "*", "*?", "+", "+?", "?", "??", "{", "{",                      \
2144    "class", "nclass", "xclass", "Ref", "Recurse", "Callout",       \    "*+","++", "?+", "{",                                           \
2145    "Alt", "Ket", "KetRmax", "KetRmin", "Assert", "Assert not",     \    "class", "nclass", "xclass", "Ref", "Refi", "DnRef", "DnRefi",  \
2146    "AssertB", "AssertB not", "Reverse",                            \    "Recurse", "Callout",                                           \
2147    "Once", "Bra", "CBra", "Cond", "SBra", "SCBra", "SCond",        \    "Alt", "Ket", "KetRmax", "KetRmin", "KetRpos",                  \
2148    "Cond ref", "Cond rec", "Cond def", "Brazero", "Braminzero",    \    "Reverse", "Assert", "Assert not", "AssertB", "AssertB not",    \
2149    "*PRUNE", "*SKIP", "*THEN", "*COMMIT", "*FAIL", "*ACCEPT"    "Once", "Once_NC",                                              \
2150      "Bra", "BraPos", "CBra", "CBraPos",                             \
2151      "Cond",                                                         \
2152      "SBra", "SBraPos", "SCBra", "SCBraPos",                         \
2153      "SCond",                                                        \
2154      "Cond ref", "Cond dnref", "Cond rec", "Cond dnrec", "Cond def", \
2155      "Brazero", "Braminzero", "Braposzero",                          \
2156      "*MARK", "*PRUNE", "*PRUNE", "*SKIP", "*SKIP",                  \
2157      "*THEN", "*THEN", "*COMMIT", "*FAIL",                           \
2158      "*ACCEPT", "*ASSERT_ACCEPT",                                    \
2159      "Close", "Skip zero"
2160    
2161    
2162  /* This macro defines the length of fixed length operations in the compiled  /* This macro defines the length of fixed length operations in the compiled
# Line 806  in UTF-8 mode. The code that uses this t Line 2172  in UTF-8 mode. The code that uses this t
2172    1,                             /* End                                    */ \    1,                             /* End                                    */ \
2173    1, 1, 1, 1, 1,                 /* \A, \G, \K, \B, \b                     */ \    1, 1, 1, 1, 1,                 /* \A, \G, \K, \B, \b                     */ \
2174    1, 1, 1, 1, 1, 1,              /* \D, \d, \S, \s, \W, \w                 */ \    1, 1, 1, 1, 1, 1,              /* \D, \d, \S, \s, \W, \w                 */ \
2175    1, 1,                          /* Any, Anybyte                           */ \    1, 1, 1,                       /* Any, AllAny, Anybyte                   */ \
2176    3, 3, 1,                       /* NOTPROP, PROP, EXTUNI                  */ \    3, 3,                          /* \P, \p                                 */ \
2177    1, 1, 1, 1, 1,                 /* \R, \H, \h, \V, \v                     */ \    1, 1, 1, 1, 1,                 /* \R, \H, \h, \V, \v                     */ \
2178    1, 1, 2, 1, 1,                 /* \Z, \z, Opt, ^, $                      */ \    1,                             /* \X                                     */ \
2179      1, 1, 1, 1, 1, 1,              /* \Z, \z, $, $M ^, ^M                    */ \
2180    2,                             /* Char  - the minimum length             */ \    2,                             /* Char  - the minimum length             */ \
2181    2,                             /* Charnc  - the minimum length           */ \    2,                             /* Chari  - the minimum length            */ \
2182    2,                             /* not                                    */ \    2,                             /* not                                    */ \
2183    /* Positive single-char repeats                            ** These are  */ \    2,                             /* noti                                   */ \
2184    2, 2, 2, 2, 2, 2,              /* *, *?, +, +?, ?, ??      ** minima in  */ \    /* Positive single-char repeats                             ** These are */ \
2185    4, 4, 4,                       /* upto, minupto, exact     ** UTF-8 mode */ \    2, 2, 2, 2, 2, 2,              /* *, *?, +, +?, ?, ??       ** minima in */ \
2186    2, 2, 2, 4,                    /* *+, ++, ?+, upto+                      */ \    2+IMM2_SIZE, 2+IMM2_SIZE,      /* upto, minupto             ** mode      */ \
2187      2+IMM2_SIZE,                   /* exact                                  */ \
2188      2, 2, 2, 2+IMM2_SIZE,          /* *+, ++, ?+, upto+                      */ \
2189      2, 2, 2, 2, 2, 2,              /* *I, *?I, +I, +?I, ?I, ??I ** UTF-8     */ \
2190      2+IMM2_SIZE, 2+IMM2_SIZE,      /* upto I, minupto I                      */ \
2191      2+IMM2_SIZE,                   /* exact I                                */ \
2192      2, 2, 2, 2+IMM2_SIZE,          /* *+I, ++I, ?+I, upto+I                  */ \
2193    /* Negative single-char repeats - only for chars < 256                   */ \    /* Negative single-char repeats - only for chars < 256                   */ \
2194    2, 2, 2, 2, 2, 2,              /* NOT *, *?, +, +?, ?, ??                */ \    2, 2, 2, 2, 2, 2,              /* NOT *, *?, +, +?, ?, ??                */ \
2195    4, 4, 4,                       /* NOT upto, minupto, exact               */ \    2+IMM2_SIZE, 2+IMM2_SIZE,      /* NOT upto, minupto                      */ \
2196    2, 2, 2, 4,                    /* Possessive *, +, ?, upto               */ \    2+IMM2_SIZE,                   /* NOT exact                              */ \
2197      2, 2, 2, 2+IMM2_SIZE,          /* Possessive NOT *, +, ?, upto           */ \
2198      2, 2, 2, 2, 2, 2,              /* NOT *I, *?I, +I, +?I, ?I, ??I          */ \
2199      2+IMM2_SIZE, 2+IMM2_SIZE,      /* NOT upto I, minupto I                  */ \
2200      2+IMM2_SIZE,                   /* NOT exact I                            */ \
2201      2, 2, 2, 2+IMM2_SIZE,          /* Possessive NOT *I, +I, ?I, upto I      */ \
2202    /* Positive type repeats                                                 */ \    /* Positive type repeats                                                 */ \
2203    2, 2, 2, 2, 2, 2,              /* Type *, *?, +, +?, ?, ??               */ \    2, 2, 2, 2, 2, 2,              /* Type *, *?, +, +?, ?, ??               */ \
2204    4, 4, 4,                       /* Type upto, minupto, exact              */ \    2+IMM2_SIZE, 2+IMM2_SIZE,      /* Type upto, minupto                     */ \
2205    2, 2, 2, 4,                    /* Possessive *+, ++, ?+, upto+           */ \    2+IMM2_SIZE,                   /* Type exact                             */ \
2206      2, 2, 2, 2+IMM2_SIZE,          /* Possessive *+, ++, ?+, upto+           */ \
2207    /* Character class & ref repeats                                         */ \    /* Character class & ref repeats                                         */ \
2208    1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */ \    1, 1, 1, 1, 1, 1,              /* *, *?, +, +?, ?, ??                    */ \
2209    5, 5,                          /* CRRANGE, CRMINRANGE                    */ \    1+2*IMM2_SIZE, 1+2*IMM2_SIZE,  /* CRRANGE, CRMINRANGE                    */ \
2210   33,                             /* CLASS                                  */ \    1, 1, 1, 1+2*IMM2_SIZE,        /* Possessive *+, ++, ?+, CRPOSRANGE      */ \
2211   33,                             /* NCLASS                                 */ \    1+(32/sizeof(pcre_uchar)),     /* CLASS                                  */ \
2212      1+(32/sizeof(pcre_uchar)),     /* NCLASS                                 */ \
2213    0,                             /* XCLASS - variable length               */ \    0,                             /* XCLASS - variable length               */ \
2214    3,                             /* REF                                    */ \    1+IMM2_SIZE,                   /* REF                                    */ \
2215      1+IMM2_SIZE,                   /* REFI                                   */ \
2216      1+2*IMM2_SIZE,                 /* DNREF                                  */ \
2217      1+2*IMM2_SIZE,                 /* DNREFI                                 */ \
2218    1+LINK_SIZE,                   /* RECURSE                                */ \    1+LINK_SIZE,                   /* RECURSE                                */ \
2219    2+2*LINK_SIZE,                 /* CALLOUT                                */ \    2+2*LINK_SIZE,                 /* CALLOUT                                */ \
2220    1+LINK_SIZE,                   /* Alt                                    */ \    1+LINK_SIZE,                   /* Alt                                    */ \
2221    1+LINK_SIZE,                   /* Ket                                    */ \    1+LINK_SIZE,                   /* Ket                                    */ \
2222    1+LINK_SIZE,                   /* KetRmax                                */ \    1+LINK_SIZE,                   /* KetRmax                                */ \
2223    1+LINK_SIZE,                   /* KetRmin                                */ \    1+LINK_SIZE,                   /* KetRmin                                */ \
2224      1+LINK_SIZE,                   /* KetRpos                                */ \
2225      1+LINK_SIZE,                   /* Reverse                                */ \
2226    1+LINK_SIZE,                   /* Assert                                 */ \    1+LINK_SIZE,                   /* Assert                                 */ \
2227    1+LINK_SIZE,                   /* Assert not                             */ \    1+LINK_SIZE,                   /* Assert not                             */ \
2228    1+LINK_SIZE,                   /* Assert behind                          */ \    1+LINK_SIZE,                   /* Assert behind                          */ \
2229    1+LINK_SIZE,                   /* Assert behind not                      */ \    1+LINK_SIZE,                   /* Assert behind not                      */ \
   1+LINK_SIZE,                   /* Reverse                                */ \  
2230    1+LINK_SIZE,                   /* ONCE                                   */ \    1+LINK_SIZE,                   /* ONCE                                   */ \
2231      1+LINK_SIZE,                   /* ONCE_NC                                */ \
2232    1+LINK_SIZE,                   /* BRA                                    */ \    1+LINK_SIZE,                   /* BRA                                    */ \
2233    3+LINK_SIZE,                   /* CBRA                                   */ \    1+LINK_SIZE,                   /* BRAPOS                                 */ \
2234      1+LINK_SIZE+IMM2_SIZE,         /* CBRA                                   */ \
2235      1+LINK_SIZE+IMM2_SIZE,         /* CBRAPOS                                */ \
2236    1+LINK_SIZE,                   /* COND                                   */ \    1+LINK_SIZE,                   /* COND                                   */ \
2237    1+LINK_SIZE,                   /* SBRA                                   */ \    1+LINK_SIZE,                   /* SBRA                                   */ \
2238    3+LINK_SIZE,                   /* SCBRA                                  */ \    1+LINK_SIZE,                   /* SBRAPOS                                */ \
2239      1+LINK_SIZE+IMM2_SIZE,         /* SCBRA                                  */ \
2240      1+LINK_SIZE+IMM2_SIZE,         /* SCBRAPOS                               */ \
2241    1+LINK_SIZE,                   /* SCOND                                  */ \    1+LINK_SIZE,                   /* SCOND                                  */ \
2242    3,                             /* CREF                                   */ \    1+IMM2_SIZE, 1+2*IMM2_SIZE,    /* CREF, DNCREF                           */ \
2243    3,                             /* RREF                                   */ \    1+IMM2_SIZE, 1+2*IMM2_SIZE,    /* RREF, DNRREF                           */ \
2244    1,                             /* DEF                                    */ \    1,                             /* DEF                                    */ \
2245    1, 1,                          /* BRAZERO, BRAMINZERO                    */ \    1, 1, 1,                       /* BRAZERO, BRAMINZERO, BRAPOSZERO        */ \
2246    1, 1, 1, 1,                    /* PRUNE, SKIP, THEN, COMMIT,             */ \    3, 1, 3,                       /* MARK, PRUNE, PRUNE_ARG                 */ \
2247    1, 1                           /* FAIL, ACCEPT                           */    1, 3,                          /* SKIP, SKIP_ARG                         */ \
2248      1, 3,                          /* THEN, THEN_ARG                         */ \
2249      1, 1, 1, 1,                    /* COMMIT, FAIL, ACCEPT, ASSERT_ACCEPT    */ \
2250      1+IMM2_SIZE, 1                 /* CLOSE, SKIPZERO                        */
2251    
2252  /* A magic value for OP_RREF to indicate the "any recursion" condition. */  /* A magic value for OP_RREF to indicate the "any recursion" condition. */
2253    
2254  #define RREF_ANY  0xffff  #define RREF_ANY  0xffff
2255    
2256  /* Error code numbers. They are given names so that they can more easily be  /* Compile time error code numbers. They are given names so that they can more
2257  tracked. */  easily be tracked. When a new number is added, the table called eint in
2258    pcreposix.c must be updated. */
2259    
2260  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,  ERR5,  ERR6,  ERR7,  ERR8,  ERR9,
2261         ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,         ERR10, ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19,
# Line 871  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4, Line 2263  enum { ERR0,  ERR1,  ERR2,  ERR3,  ERR4,
2263         ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,         ERR30, ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39,
2264         ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,         ERR40, ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47, ERR48, ERR49,
2265         ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,         ERR50, ERR51, ERR52, ERR53, ERR54, ERR55, ERR56, ERR57, ERR58, ERR59,
2266         ERR60, ERR61, ERR62 };         ERR60, ERR61, ERR62, ERR63, ERR64, ERR65, ERR66, ERR67, ERR68, ERR69,
2267           ERR70, ERR71, ERR72, ERR73, ERR74, ERR75, ERR76, ERR77, ERR78, ERR79,
2268           ERR80, ERR81, ERR82, ERR83, ERR84, ERRCOUNT };
2269    
2270    /* JIT compiling modes. The function list is indexed by them. */
2271    
2272    enum { JIT_COMPILE, JIT_PARTIAL_SOFT_COMPILE, JIT_PARTIAL_HARD_COMPILE,
2273           JIT_NUMBER_OF_COMPILE_MODES };
2274    
2275  /* The real format of the start of the pcre block; the index of names and the  /* The real format of the start of the pcre block; the index of names and the
2276  code vector run on as long as necessary after the end. We store an explicit  code vector run on as long as necessary after the end. We store an explicit
2277  offset to the name table so that if a regex is compiled on one host, saved, and  offset to the name table so that if a regex is compiled on one host, saved, and
2278  then run on another where the size of pointers is different, all might still  then run on another where the size of pointers is different, all might still
2279  be well. For the case of compiled-on-4 and run-on-8, we include an extra  be well.
2280  pointer that is always NULL. For future-proofing, a few dummy fields were  
2281  originally included - even though you can never get this planning right - but  The size of the structure must be a multiple of 8 bytes. For the case of
2282  there is only one left now.  compiled-on-4 and run-on-8, we include an extra pointer that is always NULL so
2283    that there are an even number of pointers which therefore are a multiple of 8
2284  NOTE NOTE NOTE:  bytes.
2285  Because people can now save and re-use compiled patterns, any additions to this  
2286  structure should be made at the end, and something earlier (e.g. a new  It is necessary to fork the struct for the 32 bit library, since it needs to
2287  flag in the options or one of the dummy fields) should indicate that the new  use pcre_uint32 for first_char and req_char. We can't put an ifdef inside the
2288  fields are present. Currently PCRE always sets the dummy fields to zero.  typedef because pcretest needs access to the struct of the 8-, 16- and 32-bit
2289  NOTE NOTE NOTE:  variants.
2290    
2291    *** WARNING ***
2292    When new fields are added to these structures, remember to adjust the code in
2293    pcre_byte_order.c that is concerned with swapping the byte order of the fields
2294    when a compiled regex is reloaded on a host with different endianness.
2295    *** WARNING ***
2296    There is also similar byte-flipping code in pcretest.c, which is used for
2297    testing the byte-flipping features. It must also be kept in step.
2298    *** WARNING ***
2299  */  */
2300    
2301  typedef struct real_pcre {  typedef struct real_pcre8_or_16 {
2302      pcre_uint32 magic_number;
2303      pcre_uint32 size;               /* Total that was malloced */
2304      pcre_uint32 options;            /* Public options */
2305      pcre_uint32 flags;              /* Private flags */
2306      pcre_uint32 limit_match;        /* Limit set from regex */
2307      pcre_uint32 limit_recursion;    /* Limit set from regex */
2308      pcre_uint16 first_char;         /* Starting character */
2309      pcre_uint16 req_char;           /* This character must be seen */
2310      pcre_uint16 max_lookbehind;     /* Longest lookbehind (characters) */
2311      pcre_uint16 top_bracket;        /* Highest numbered group */
2312      pcre_uint16 top_backref;        /* Highest numbered back reference */
2313      pcre_uint16 name_table_offset;  /* Offset to name table that follows */
2314      pcre_uint16 name_entry_size;    /* Size of any name items */
2315      pcre_uint16 name_count;         /* Number of name items */
2316      pcre_uint16 ref_count;          /* Reference count */
2317      pcre_uint16 dummy1;             /* To ensure size is a multiple of 8 */
2318      pcre_uint16 dummy2;             /* To ensure size is a multiple of 8 */
2319      pcre_uint16 dummy3;             /* To ensure size is a multiple of 8 */
2320      const pcre_uint8 *tables;       /* Pointer to tables or NULL for std */
2321      void             *nullpad;      /* NULL padding */
2322    } real_pcre8_or_16;
2323    
2324    typedef struct real_pcre8_or_16 real_pcre;
2325    typedef struct real_pcre8_or_16 real_pcre16;
2326    
2327    typedef struct real_pcre32 {
2328    pcre_uint32 magic_number;    pcre_uint32 magic_number;
2329    pcre_uint32 size;               /* Total that was malloced */    pcre_uint32 size;               /* Total that was malloced */
2330    pcre_uint32 options;            /* Public options */    pcre_uint32 options;            /* Public options */
2331    pcre_uint16 flags;              /* Private flags */    pcre_uint32 flags;              /* Private flags */
2332    pcre_uint16 dummy1;             /* For future use */    pcre_uint32 limit_match;        /* Limit set from regex */
2333    pcre_uint16 top_bracket;    pcre_uint32 limit_recursion;    /* Limit set from regex */
2334    pcre_uint16 top_backref;    pcre_uint32 first_char;         /* Starting character */
2335    pcre_uint16 first_byte;    pcre_uint32 req_char;           /* This character must be seen */
2336    pcre_uint16 req_byte;    pcre_uint16 max_lookbehind;     /* Longest lookbehind (characters) */
2337      pcre_uint16 top_bracket;        /* Highest numbered group */
2338      pcre_uint16 top_backref;        /* Highest numbered back reference */
2339    pcre_uint16 name_table_offset;  /* Offset to name table that follows */    pcre_uint16 name_table_offset;  /* Offset to name table that follows */
2340    pcre_uint16 name_entry_size;    /* Size of any name items */    pcre_uint16 name_entry_size;    /* Size of any name items */
2341    pcre_uint16 name_count;         /* Number of name items */    pcre_uint16 name_count;         /* Number of name items */
2342    pcre_uint16 ref_count;          /* Reference count */    pcre_uint16 ref_count;          /* Reference count */
2343      pcre_uint16 dummy;              /* To ensure size is a multiple of 8 */
2344      const pcre_uint8 *tables;       /* Pointer to tables or NULL for std */
2345      void             *nullpad;      /* NULL padding */
2346    } real_pcre32;
2347    
2348    #if defined COMPILE_PCRE8
2349    #define REAL_PCRE real_pcre
2350    #elif defined COMPILE_PCRE16
2351    #define REAL_PCRE real_pcre16
2352    #elif defined COMPILE_PCRE32
2353    #define REAL_PCRE real_pcre32
2354    #endif
2355    
2356    /* Assert that the size of REAL_PCRE is divisible by 8 */
2357    typedef int __assert_real_pcre_size_divisible_8[(sizeof(REAL_PCRE) % 8) == 0 ? 1 : -1];
2358    
2359    const unsigned char *tables;    /* Pointer to tables or NULL for std */  /* Needed in pcretest to access some fields in the real_pcre* structures
2360    const unsigned char *nullpad;   /* NULL padding */   * directly. They're unified for 8/16/32 bits since the structs only differ
2361  } real_pcre;   * after these fields; if that ever changes, need to fork those defines into
2362     * 8/16 and 32 bit versions. */
2363    #define REAL_PCRE_MAGIC(re)     (((REAL_PCRE*)re)->magic_number)
2364    #define REAL_PCRE_SIZE(re)      (((REAL_PCRE*)re)->size)
2365    #define REAL_PCRE_OPTIONS(re)   (((REAL_PCRE*)re)->options)
2366    #define REAL_PCRE_FLAGS(re)     (((REAL_PCRE*)re)->flags)
2367    
2368  /* The format of the block used to store data from pcre_study(). The same  /* The format of the block used to store data from pcre_study(). The same
2369  remark (see NOTE above) about extending this structure applies. */  remark (see NOTE above) about extending this structure applies. */
2370    
2371  typedef struct pcre_study_data {  typedef struct pcre_study_data {
2372    pcre_uint32 size;               /* Total that was malloced */    pcre_uint32 size;               /* Total that was malloced */
2373    pcre_uint32 options;    pcre_uint32 flags;              /* Private flags */
2374    uschar start_bits[32];    pcre_uint8 start_bits[32];      /* Starting char bits */
2375      pcre_uint32 minlength;          /* Minimum subject length */
2376  } pcre_study_data;  } pcre_study_data;
2377    
2378    /* Structure for building a chain of open capturing subpatterns during
2379    compiling, so that instructions to close them can be compiled when (*ACCEPT) is
2380    encountered. This is also used to identify subpatterns that contain recursive
2381    back references to themselves, so that they can be made atomic. */
2382    
2383    typedef struct open_capitem {
2384      struct open_capitem *next;    /* Chain link */
2385      pcre_uint16 number;           /* Capture number */
2386      pcre_uint16 flag;             /* Set TRUE if recursive back ref */
2387    } open_capitem;
2388    
2389    /* Structure for building a list of named groups during the first pass of
2390    compiling. */
2391    
2392    typedef struct named_group {
2393      const pcre_uchar  *name;          /* Points to the name in the pattern */
2394      int                length;        /* Length of the name */
2395      pcre_uint32        number;        /* Group number */
2396    } named_group;
2397    
2398  /* Structure for passing "static" information around between the functions  /* Structure for passing "static" information around between the functions
2399  doing the compiling, so that they are thread-safe. */  doing the compiling, so that they are thread-safe. */
2400    
2401  typedef struct compile_data {  typedef struct compile_data {
2402    const uschar *lcc;            /* Points to lower casing table */    const pcre_uint8 *lcc;            /* Points to lower casing table */
2403    const uschar *fcc;            /* Points to case-flipping table */    const pcre_uint8 *fcc;            /* Points to case-flipping table */
2404    const uschar *cbits;          /* Points to character type table */    const pcre_uint8 *cbits;          /* Points to character type table */
2405    const uschar *ctypes;         /* Points to table of type maps */    const pcre_uint8 *ctypes;         /* Points to table of type maps */
2406    const uschar *start_workspace;/* The start of working space */    const pcre_uchar *start_workspace;/* The start of working space */
2407    const uschar *start_code;     /* The start of the compiled code */    const pcre_uchar *start_code;     /* The start of the compiled code */
2408    const uschar *start_pattern;  /* The start of the pattern */    const pcre_uchar *start_pattern;  /* The start of the pattern */
2409    const uschar *end_pattern;    /* The end of the pattern */    const pcre_uchar *end_pattern;    /* The end of the pattern */
2410    uschar *hwm;                  /* High watermark of workspace */    pcre_uchar *hwm;                  /* High watermark of workspace */
2411    uschar *name_table;           /* The name/number table */    open_capitem *open_caps;          /* Chain of open capture items */
2412    int  names_found;             /* Number of entries so far */    named_group *named_groups;        /* Points to vector in pre-compile */
2413    int  name_entry_size;         /* Size of each entry */    pcre_uchar *name_table;           /* The name/number table */
2414    int  bracount;                /* Count of capturing parens */    int  names_found;                 /* Number of entries so far */
2415    int  top_backref;             /* Maximum back reference */    int  name_entry_size;             /* Size of each entry */
2416    unsigned int backref_map;     /* Bitmap of low back refs */    int  named_group_list_size;       /* Number of entries in the list */
2417    int  external_options;        /* External (initial) options */    int  workspace_size;              /* Size of workspace */
2418    int  external_flags;          /* External flag bits to be set */    unsigned int bracount;            /* Count of capturing parens as we compile */
2419    int  req_varyopt;             /* "After variable item" flag for reqbyte */    int  final_bracount;              /* Saved value after first pass */
2420    BOOL had_accept;              /* (*ACCEPT) encountered */    int  max_lookbehind;              /* Maximum lookbehind (characters) */
2421    int  nltype;                  /* Newline type */    int  top_backref;                 /* Maximum back reference */
2422    int  nllen;                   /* Newline string length */    unsigned int backref_map;         /* Bitmap of low back refs */
2423    uschar nl[4];                 /* Newline string when fixed length */    unsigned int namedrefcount;       /* Number of backreferences by name */
2424      int  parens_depth;                /* Depth of nested parentheses */
2425      int  assert_depth;                /* Depth of nested assertions */
2426      pcre_uint32 external_options;     /* External (initial) options */
2427      pcre_uint32 external_flags;       /* External flag bits to be set */
2428      int  req_varyopt;                 /* "After variable item" flag for reqbyte */
2429      BOOL had_accept;                  /* (*ACCEPT) encountered */
2430      BOOL had_pruneorskip;             /* (*PRUNE) or (*SKIP) encountered */
2431      BOOL check_lookbehind;            /* Lookbehinds need later checking */
2432      BOOL dupnames;                    /* Duplicate names exist */
2433      int  nltype;                      /* Newline type */
2434      int  nllen;                       /* Newline string length */
2435      pcre_uchar nl[4];                 /* Newline string when fixed length */
2436  } compile_data;  } compile_data;
2437    
2438  /* Structure for maintaining a chain of pointers to the currently incomplete  /* Structure for maintaining a chain of pointers to the currently incomplete
2439  branches, for testing for left recursion. */  branches, for testing for left recursion while compiling. */
2440    
2441  typedef struct branch_chain {  typedef struct branch_chain {
2442    struct branch_chain *outer;    struct branch_chain *outer;
2443    uschar *current;    pcre_uchar *current_branch;
2444  } branch_chain;  } branch_chain;
2445    
2446  /* Structure for items in a linked list that represents an explicit recursive  /* Structure for items in a linked list that represents an explicit recursive
2447  call within the pattern. */  call within the pattern; used by pcre_exec(). */
2448    
2449  typedef struct recursion_info {  typedef struct recursion_info {
2450    struct recursion_info *prevrec; /* Previous recursion record (or NULL) */    struct recursion_info *prevrec; /* Previous recursion record (or NULL) */
2451    int group_num;                /* Number of group that was called */    unsigned int group_num;         /* Number of group that was called */
2452    const uschar *after_call;     /* "Return value": points after the call in the expr */    int *offset_save;               /* Pointer to start of saved offsets */
2453    USPTR save_start;             /* Old value of mstart */    int saved_max;                  /* Number of saved offsets */
2454    int *offset_save;             /* Pointer to start of saved offsets */    int saved_capture_last;         /* Last capture number */
2455    int saved_max;                /* Number of saved offsets */    PCRE_PUCHAR subject_position;   /* Position at start of recursion */
2456  } recursion_info;  } recursion_info;
2457    
2458    /* A similar structure for pcre_dfa_exec(). */
2459    
2460    typedef struct dfa_recursion_info {
2461      struct dfa_recursion_info *prevrec;
2462      int group_num;
2463      PCRE_PUCHAR subject_position;
2464    } dfa_recursion_info;
2465    
2466  /* Structure for building a chain of data for holding the values of the subject  /* Structure for building a chain of data for holding the values of the subject
2467  pointer at the start of each subpattern, so as to detect when an empty string  pointer at the start of each subpattern, so as to detect when an empty string
2468  has been matched by a subpattern - to break infinite loops. */  has been matched by a subpattern - to break infinite loops; used by
2469    pcre_exec(). */
2470    
2471  typedef struct eptrblock {  typedef struct eptrblock {
2472    struct eptrblock *epb_prev;    struct eptrblock *epb_prev;
2473    USPTR epb_saved_eptr;    PCRE_PUCHAR epb_saved_eptr;
2474  } eptrblock;  } eptrblock;
2475    
2476    
# Line 983  typedef struct match_data { Line 2481  typedef struct match_data {
2481    unsigned long int match_call_count;      /* As it says */    unsigned long int match_call_count;      /* As it says */
2482    unsigned long int match_limit;           /* As it says */    unsigned long int match_limit;           /* As it says */
2483    unsigned long int match_limit_recursion; /* As it says */    unsigned long int match_limit_recursion; /* As it says */
2484    int   *offset_vector;         /* Offset vector */    int   *offset_vector;           /* Offset vector */
2485    int    offset_end;            /* One past the end */    int    offset_end;              /* One past the end */
2486    int    offset_max;            /* The maximum usable for return data */    int    offset_max;              /* The maximum usable for return data */
2487    int    nltype;                /* Newline type */    int    nltype;                  /* Newline type */
2488    int    nllen;                 /* Newline string length */    int    nllen;                   /* Newline string length */
2489    uschar nl[4];                 /* Newline string when fixed */    int    name_count;              /* Number of names in name table */
2490    const uschar *lcc;            /* Points to lower casing table */    int    name_entry_size;         /* Size of entry in names table */
2491    const uschar *ctypes;         /* Points to table of type maps */    unsigned int skip_arg_count;    /* For counting SKIP_ARGs */
2492    BOOL   offset_overflow;       /* Set if too many extractions */    unsigned int ignore_skip_arg;   /* For re-run when SKIP arg name not found */
2493    BOOL   notbol;                /* NOTBOL flag */    pcre_uchar *name_table;         /* Table of names */
2494    BOOL   noteol;                /* NOTEOL flag */    pcre_uchar nl[4];               /* Newline string when fixed */
2495    BOOL   utf8;                  /* UTF8 flag */    const  pcre_uint8 *lcc;         /* Points to lower casing table */
2496    BOOL   endonly;               /* Dollar not before final \n */    const  pcre_uint8 *fcc;         /* Points to case-flipping table */
2497    BOOL   notempty;              /* Empty string match not wanted */    const  pcre_uint8 *ctypes;      /* Points to table of type maps */
2498    BOOL   partial;               /* PARTIAL flag */    BOOL   notbol;                  /* NOTBOL flag */
2499    BOOL   hitend;                /* Hit the end of the subject at some point */    BOOL   noteol;                  /* NOTEOL flag */
2500    BOOL   bsr_anycrlf;           /* \R is just any CRLF, not full Unicode */    BOOL   utf;                     /* UTF-8 / UTF-16 flag */
2501    const uschar *start_code;     /* For use when recursing */    BOOL   jscript_compat;          /* JAVASCRIPT_COMPAT flag */
2502    USPTR  start_subject;         /* Start of the subject string */    BOOL   use_ucp;                 /* PCRE_UCP flag */
2503    USPTR  end_subject;           /* End of the subject string */    BOOL   endonly;                 /* Dollar not before final \n */
2504    USPTR  start_match_ptr;       /* Start of matched string */    BOOL   notempty;                /* Empty string match not wanted */
2505    USPTR  end_match_ptr;         /* Subject position at end match */    BOOL   notempty_atstart;        /* Empty string match at start not wanted */
2506    int    end_offset_top;        /* Highwater mark at end of match */    BOOL   hitend;                  /* Hit the end of the subject at some point */
2507    int    capture_last;          /* Most recent capture number */    BOOL   bsr_anycrlf;             /* \R is just any CRLF, not full Unicode */
2508    int    start_offset;          /* The start offset value */    BOOL   hasthen;                 /* Pattern contains (*THEN) */
2509    eptrblock *eptrchain;         /* Chain of eptrblocks for tail recursions */    const  pcre_uchar *start_code;  /* For use when recursing */
2510    int    eptrn;                 /* Next free eptrblock */    PCRE_PUCHAR start_subject;      /* Start of the subject string */
2511    recursion_info *recursive;    /* Linked list of recursion data */    PCRE_PUCHAR end_subject;        /* End of the subject string */
2512    void  *callout_data;          /* To pass back to callouts */    PCRE_PUCHAR start_match_ptr;    /* Start of matched string */
2513      PCRE_PUCHAR end_match_ptr;      /* Subject position at end match */
2514      PCRE_PUCHAR start_used_ptr;     /* Earliest consulted character */
2515      int    partial;                 /* PARTIAL options */
2516      int    end_offset_top;          /* Highwater mark at end of match */
2517      pcre_int32 capture_last;        /* Most recent capture number + overflow flag */
2518      int    start_offset;            /* The start offset value */
2519      int    match_function_type;     /* Set for certain special calls of MATCH() */
2520      eptrblock *eptrchain;           /* Chain of eptrblocks for tail recursions */
2521      int    eptrn;                   /* Next free eptrblock */
2522      recursion_info *recursive;      /* Linked list of recursion data */
2523      void  *callout_data;            /* To pass back to callouts */
2524      const  pcre_uchar *mark;        /* Mark pointer to pass back on success */
2525      const  pcre_uchar *nomatch_mark;/* Mark pointer to pass back on failure */
2526      const  pcre_uchar *once_target; /* Where to back up to for atomic groups */
2527    #ifdef NO_RECURSE
2528      void  *match_frames_base;       /* For remembering malloc'd frames */
2529    #endif
2530  } match_data;  } match_data;
2531    
2532  /* A similar structure is used for the same purpose by the DFA matching  /* A similar structure is used for the same purpose by the DFA matching
2533  functions. */  functions. */
2534    
2535  typedef struct dfa_match_data {  typedef struct dfa_match_data {
2536    const uschar *start_code;     /* Start of the compiled pattern */    const pcre_uchar *start_code;     /* Start of the compiled pattern */
2537    const uschar *start_subject;  /* Start of the subject string */    const pcre_uchar *start_subject ; /* Start of the subject string */
2538    const uschar *end_subject;    /* End of subject string */    const pcre_uchar *end_subject;    /* End of subject string */
2539    const uschar *tables;         /* Character tables */    const pcre_uchar *start_used_ptr; /* Earliest consulted character */
2540    int   moptions;               /* Match options */    const pcre_uint8 *tables;         /* Character tables */
2541    int   poptions;               /* Pattern options */    int   start_offset;               /* The start offset value */
2542    int    nltype;                /* Newline type */    int   moptions;                   /* Match options */
2543    int    nllen;                 /* Newline string length */    int   poptions;                   /* Pattern options */
2544    uschar nl[4];                 /* Newline string when fixed */    int   nltype;                     /* Newline type */
2545    void  *callout_data;          /* To pass back to callouts */    int   nllen;                      /* Newline string length */
2546      pcre_uchar nl[4];                 /* Newline string when fixed */
2547      void *callout_data;               /* To pass back to callouts */
2548      dfa_recursion_info *recursive;    /* Linked list of recursion data */
2549  } dfa_match_data;  } dfa_match_data;
2550    
2551  /* Bit definitions for entries in the pcre_ctypes table. */  /* Bit definitions for entries in the pcre_ctypes table. */
# Line 1036  typedef struct dfa_match_data { Line 2554  typedef struct dfa_match_data {
2554  #define ctype_letter  0x02  #define ctype_letter  0x02
2555  #define ctype_digit   0x04  #define ctype_digit   0x04
2556  #define ctype_xdigit  0x08  #define ctype_xdigit  0x08
2557  #define ctype_word    0x10   /* alphameric or '_' */  #define ctype_word    0x10   /* alphanumeric or '_' */
2558  #define ctype_meta    0x80   /* regexp meta char or zero (end pattern) */  #define ctype_meta    0x80   /* regexp meta char or zero (end pattern) */
2559    
2560  /* Offsets for the bitmap tables in pcre_cbits. Each table contains a set  /* Offsets for the bitmap tables in pcre_cbits. Each table contains a set
# Line 1063  total length. */ Line 2581  total length. */
2581  #define ctypes_offset (cbits_offset + cbit_length)  #define ctypes_offset (cbits_offset + cbit_length)
2582  #define tables_length (ctypes_offset + 256)  #define tables_length (ctypes_offset + 256)
2583    
2584    /* Internal function and data prefixes. */
2585    
2586    #if defined COMPILE_PCRE8
2587    #ifndef PUBL
2588    #define PUBL(name) pcre_##name
2589    #endif
2590    #ifndef PRIV
2591    #define PRIV(name) _pcre_##name
2592    #endif
2593    #elif defined COMPILE_PCRE16
2594    #ifndef PUBL
2595    #define PUBL(name) pcre16_##name
2596    #endif
2597    #ifndef PRIV
2598    #define PRIV(name) _pcre16_##name
2599    #endif
2600    #elif defined COMPILE_PCRE32
2601    #ifndef PUBL
2602    #define PUBL(name) pcre32_##name
2603    #endif
2604    #ifndef PRIV
2605    #define PRIV(name) _pcre32_##name
2606    #endif
2607    #else
2608    #error Unsupported compiling mode
2609    #endif /* COMPILE_PCRE[8|16|32] */
2610    
2611  /* Layout of the UCP type table that translates property names into types and  /* Layout of the UCP type table that translates property names into types and
2612  codes. Each entry used to point directly to a name, but to reduce the number of  codes. Each entry used to point directly to a name, but to reduce the number of
2613  relocations in shared libraries, it now has an offset into a single string  relocations in shared libraries, it now has an offset into a single string
# Line 1080  of the exported public functions. They h Line 2625  of the exported public functions. They h
2625  but are not part of the PCRE public API. The data for these tables is in the  but are not part of the PCRE public API. The data for these tables is in the
2626  pcre_tables.c module. */  pcre_tables.c module. */
2627    
2628  extern const int    _pcre_utf8_table1[];  #ifdef COMPILE_PCRE8
2629  extern const int    _pcre_utf8_table2[];  extern const int            PRIV(utf8_table1)[];
2630  extern const int    _pcre_utf8_table3[];  extern const int            PRIV(utf8_table1_size);
2631  extern const uschar _pcre_utf8_table4[];  extern const int            PRIV(utf8_table2)[];
2632    extern const int            PRIV(utf8_table3)[];
2633  extern const int    _pcre_utf8_table1_size;  extern const pcre_uint8     PRIV(utf8_table4)[];
2634    #endif /* COMPILE_PCRE8 */
2635  extern const char   _pcre_utt_names[];  
2636  extern const ucp_type_table _pcre_utt[];  extern const char           PRIV(utt_names)[];
2637  extern const int _pcre_utt_size;  extern const ucp_type_table PRIV(utt)[];
2638    extern const int            PRIV(utt_size);
2639    
2640  extern const uschar _pcre_default_tables[];  extern const pcre_uint8     PRIV(OP_lengths)[];
2641    extern const pcre_uint8     PRIV(default_tables)[];
2642    
2643  extern const uschar _pcre_OP_lengths[];  extern const pcre_uint32    PRIV(hspace_list)[];
2644    extern const pcre_uint32    PRIV(vspace_list)[];
2645    
2646    
2647  /* Internal shared functions. These are functions that are used by more than  /* Internal shared functions. These are functions that are used by more than
2648  one of the exported public functions. They have to be "external" in the C  one of the exported public functions. They have to be "external" in the C
2649  sense, but are not part of the PCRE public API. */  sense, but are not part of the PCRE public API. */
2650    
2651  extern BOOL         _pcre_is_newline(const uschar *, int, const uschar *,  /* String comparison functions. */
2652                        int *, BOOL);  #if defined COMPILE_PCRE8
2653  extern int          _pcre_ord2utf8(int, uschar *);  
2654  extern real_pcre   *_pcre_try_flipped(const real_pcre *, real_pcre *,  #define STRCMP_UC_UC(str1, str2) \
2655                        const pcre_study_data *, pcre_study_data *);    strcmp((char *)(str1), (char *)(str2))
2656  extern int          _pcre_ucp_findprop(const unsigned int, int *, int *);  #define STRCMP_UC_C8(str1, str2) \
2657  extern unsigned int _pcre_ucp_othercase(const unsigned int);    strcmp((char *)(str1), (str2))
2658  extern int          _pcre_valid_utf8(const uschar *, int);  #define STRNCMP_UC_UC(str1, str2, num) \
2659  extern BOOL         _pcre_was_newline(const uschar *, int, const uschar *,    strncmp((char *)(str1), (char *)(str2), (num))
2660                        int *, BOOL);  #define STRNCMP_UC_C8(str1, str2, num) \
2661  extern BOOL         _pcre_xclass(int, const uschar *);    strncmp((char *)(str1), (str2), (num))
2662    #define STRLEN_UC(str) strlen((const char *)str)
2663    
2664    #elif defined COMPILE_PCRE16 || defined COMPILE_PCRE32
2665    
2666    extern int               PRIV(strcmp_uc_uc)(const pcre_uchar *,
2667                               const pcre_uchar *);
2668    extern int               PRIV(strcmp_uc_c8)(const pcre_uchar *,
2669                               const char *);
2670    extern int               PRIV(strncmp_uc_uc)(const pcre_uchar *,
2671                               const pcre_uchar *, unsigned int num);
2672    extern int               PRIV(strncmp_uc_c8)(const pcre_uchar *,
2673                               const char *, unsigned int num);
2674    extern unsigned int      PRIV(strlen_uc)(const pcre_uchar *str);
2675    
2676    #define STRCMP_UC_UC(str1, str2) \
2677      PRIV(strcmp_uc_uc)((str1), (str2))
2678    #define STRCMP_UC_C8(str1, str2) \
2679      PRIV(strcmp_uc_c8)((str1), (str2))
2680    #define STRNCMP_UC_UC(str1, str2, num) \
2681      PRIV(strncmp_uc_uc)((str1), (str2), (num))
2682    #define STRNCMP_UC_C8(str1, str2, num) \
2683      PRIV(strncmp_uc_c8)((str1), (str2), (num))
2684    #define STRLEN_UC(str) PRIV(strlen_uc)(str)
2685    
2686    #endif /* COMPILE_PCRE[8|16|32] */
2687    
2688    #if defined COMPILE_PCRE8 || defined COMPILE_PCRE16
2689    
2690    #define STRCMP_UC_UC_TEST(str1, str2) STRCMP_UC_UC(str1, str2)
2691    #define STRCMP_UC_C8_TEST(str1, str2) STRCMP_UC_C8(str1, str2)
2692    
2693    #elif defined COMPILE_PCRE32
2694    
2695    extern int               PRIV(strcmp_uc_uc_utf)(const pcre_uchar *,
2696                               const pcre_uchar *);
2697    extern int               PRIV(strcmp_uc_c8_utf)(const pcre_uchar *,
2698                               const char *);
2699    
2700    #define STRCMP_UC_UC_TEST(str1, str2) \
2701      (utf ? PRIV(strcmp_uc_uc_utf)((str1), (str2)) : PRIV(strcmp_uc_uc)((str1), (str2)))
2702    #define STRCMP_UC_C8_TEST(str1, str2) \
2703      (utf ? PRIV(strcmp_uc_c8_utf)((str1), (str2)) : PRIV(strcmp_uc_c8)((str1), (str2)))
2704    
2705    #endif /* COMPILE_PCRE[8|16|32] */
2706    
2707    extern const pcre_uchar *PRIV(find_bracket)(const pcre_uchar *, BOOL, int);
2708    extern BOOL              PRIV(is_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2709                               int *, BOOL);
2710    extern unsigned int      PRIV(ord2utf)(pcre_uint32, pcre_uchar *);
2711    extern int               PRIV(valid_utf)(PCRE_PUCHAR, int, int *);
2712    extern BOOL              PRIV(was_newline)(PCRE_PUCHAR, int, PCRE_PUCHAR,
2713                               int *, BOOL);
2714    extern BOOL              PRIV(xclass)(pcre_uint32, const pcre_uchar *, BOOL);
2715    
2716    #ifdef SUPPORT_JIT
2717    extern void              PRIV(jit_compile)(const REAL_PCRE *,
2718                               PUBL(extra) *, int);
2719    extern int               PRIV(jit_exec)(const PUBL(extra) *,
2720                               const pcre_uchar *, int, int, int, int *, int);
2721    extern void              PRIV(jit_free)(void *);
2722    extern int               PRIV(jit_get_size)(void *);
2723    extern const char*       PRIV(jit_get_target)(void);
2724    #endif
2725    
2726    /* Unicode character database (UCD) */
2727    
2728    typedef struct {
2729      pcre_uint8 script;     /* ucp_Arabic, etc. */
2730      pcre_uint8 chartype;   /* ucp_Cc, etc. (general categories) */
2731      pcre_uint8 gbprop;     /* ucp_gbControl, etc. (grapheme break property) */
2732      pcre_uint8 caseset;    /* offset to multichar other cases or zero */
2733      pcre_int32 other_case; /* offset to other case, or zero if none */
2734    } ucd_record;
2735    
2736    extern const pcre_uint32 PRIV(ucd_caseless_sets)[];
2737    extern const ucd_record  PRIV(ucd_records)[];
2738    extern const pcre_uint8  PRIV(ucd_stage1)[];
2739    extern const pcre_uint16 PRIV(ucd_stage2)[];
2740    extern const pcre_uint32 PRIV(ucp_gentype)[];
2741    extern const pcre_uint32 PRIV(ucp_gbtable)[];
2742    #ifdef SUPPORT_JIT
2743    extern const int         PRIV(ucp_typerange)[];
2744    #endif
2745    
2746    #ifdef SUPPORT_UCP
2747    /* UCD access macros */
2748    
2749    #define UCD_BLOCK_SIZE 128
2750    #define GET_UCD(ch) (PRIV(ucd_records) + \
2751            PRIV(ucd_stage2)[PRIV(ucd_stage1)[(int)(ch) / UCD_BLOCK_SIZE] * \
2752            UCD_BLOCK_SIZE + (int)(ch) % UCD_BLOCK_SIZE])
2753    
2754    #define UCD_CHARTYPE(ch)    GET_UCD(ch)->chartype
2755    #define UCD_SCRIPT(ch)      GET_UCD(ch)->script
2756    #define UCD_CATEGORY(ch)    PRIV(ucp_gentype)[UCD_CHARTYPE(ch)]
2757    #define UCD_GRAPHBREAK(ch)  GET_UCD(ch)->gbprop
2758    #define UCD_CASESET(ch)     GET_UCD(ch)->caseset
2759    #define UCD_OTHERCASE(ch)   ((pcre_uint32)((int)ch + (int)(GET_UCD(ch)->other_case)))
2760    
2761    #endif /* SUPPORT_UCP */
2762    
2763  #endif  #endif
2764    

Legend:
Removed from v.268  
changed lines
  Added in v.1425

  ViewVC Help
Powered by ViewVC 1.1.5